(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Assyrian historiography, a source study"

I 




THE 



UNIVERSITYOF MISSOURI 
STUDIES 



SOCIAL SCIENCE SERIES 

-^ VOIATME III NuMnKK 1 

ASSYRIAN HISTORIOGRAPHY 

A Source Sfndv 



By 
ALBERT TEN EYCK OLMSTEAD 
Associate Professor of Ancient fiistory 




;-: OF t: 
N I V E I- 



UNIVERSITY OF MISSOURI 

Columbia, Misaouui 
May, 1916 



ASSYRIAN HISTORIOGRAPHY 
A SOURCE STUDY 






THE 



UNIVERSITYOF MISSOURI 
STUDIES 



SOCIAL SCIENCE SERIES 

Volume III Number 1 

ASSYRIAN HISTORIOGRAPHY 

A Source Study 



By 
ALBERT TEN EYCK OLMSTEAD 
Associate Professor of Ancient History 




UNIVERSITY OF MISSOURI 
Columbia, Missouri 

May, 1916 






EXCHANGE 



CONTENTS 

CHAPTER I PAGE 
Assyrian Historians and their Histories 1 

CHAPTER H 
The Beginnings of True History 

(Tiglath Pileser I) 10 

CHAPTER HI 
The Development of Historical Writing 

(Ashur nasir apal and Shalmaneser HI) 15 

CHAPTER IV 
Shamshi Adad and the Synchronistic History 29 

CHAPTER V 
Sargon and the Modern Historical Criticism 36 

CHAPTER VI 
Annals and Display Inscriptions 

(Sennacherib and Esarhaddon) 43 

CHAPTER VII 
Ashur bani apal and Assyrian Editing 53 

CHAPTER VIII 
The Babylonian Chronicle and Berossus 59 



Vll 



CHAPTER I 

ASSYRIAN HISTORIANS AND THEIR HISTORIES 

To the serious student of Assyrian history, it is obvious 
that we cannot write that history until we have adequately dis- 
cussed the sources. We must learn what these are, in other 
words, we must begin with a bibliography of the various docu- 
ments. Then we must divide them into their various classes, 
for different classes of inscriptions are of varying degrees of 
accuracy. Finally, we must study in detail for each reign the 
sources, discover which of the various documents or groups of 
documents are the most nearly contemporaneous with the events 
they narrate, and on these, and on these alone, base our history 
of the period. 

To the less narrowly technical reader, the development of 
the historical sense in one of the earlier culture peoples has an 
interest all its own. The historical writings of the Assyrians 
form one of the most important branches of their literature. In- 
deed, it may be claimed with much truth that it is the most char- 
acteristically Assyrian of them all.^ 

iThis study is a source investigation and not a bibliography. The 
only royal inscriptions studied in detail are those presenting source 
problems. Minor inscriptions of these rulers are accorded no more 
space than is absolutely necessary, and rulers who have not given 
us strictly historical inscriptions are generally passed in silence. The 
bibliographical notes are condensed as much as possible and make 
no pretense of completeness, though they will probably be found the 
most complete yet printed. Every possible care has been taken to 
make the references accurate, but the fact that many were consulted in 
the libraries of Cornell University, University of Chicago, Columbia 
University, and the University of Pennsylvania, and are thus inac- 
cessible at the time when the work is passing through the press, 
leaves some possibility of error. Dr. B. B. Charles, Instructor in 

(1) 



2 UNIVERSITY OF MISSOURI STUDIES 

The Assyrians derived their historical writing, as they did 
so many other cultural elements, from the Babylonians. In that 
country, there had existed from the earliest times two types of 
historical inscriptions. The more common form developed from 
the desire of the kings to commemorate, not their deeds in war, 
but their building operations, and more especially the buildings 
erected in honor of the gods. Now and then we have an inci- 
dental reference to military activities, but rarely indeed do we 
find a document devoted primarily to the narration of warlike 
deeds. Side by side with these building inscriptions were to be 
found dry lists of kings, sometimes with the length of their 
reigns, but, save for an occasional legend, there seem to have 
been no detailed histories. It was from the former type that the 
earliest Assyrian inscriptions were derived. In actual fact, we 
have no right to call them historical in any sense of the word, 
even though they are our only sources for the few facts we know 
about this early period. A typical inscription of this type will 
have the form "Irishum the vice gerent of the god Ashur, the 
son of Ilnshuma the vice gerent of the god Ashur, unto the 
god Ashur, his Lord, for his own life and for the life of his son 
has dedicated". Thus there was as yet little difference in form 
from their Babylonian models and the historical data were of 
the slightest. This type persisted until the latest days of the 
Assyrian empire in the inscriptions placed on the bricks, or, in 
slightly more developed form, in the inscriptions written on the 
slabs of stone used for the adornment of palace or temple. For 
these later periods, they rarely have a value other than for the 

Semitics in the University of Pennsylvania, has kindly verified those 
where error has eeemed at all likely. — For the English speaking reader, 
practically all the inscriptions for the earlier half of the history are 
found in Budge-Kjng, Annals of the Kings of Assyria. 1. For the re- 
mainder, Harper, Assyrian and Babylonian Literature, is adequate, 
though somewhat out of date. Rogers, Cuneiform Parallels to the 
Old Testament, gives an up to date translation of those passages which 
throw light on the Biblical writings. Other works cited are generally 
of interest only to specialists and the most common are cited by ao- 
breviations which will be found at the close of the study. 



ASSYRIAN HISTORIOGRAPHY 6 

architectural history, and so demand no further study in this 
place. Nevertheless, the architectural origin of the historical in- 
scription should not be forgotten. Even to the end, it is a rare J 
document which does not have as its conclusion a more or less 
full account of the building operations carried on by the monarch 
who erected it. 

It was not long until the inscriptions were incised on lime- 
stone. These slabs, giving more surface for the writing, easily 
induced the addition of other data, including naturally some ac- 
count of the monarch's exploits in war. The typical inscription 
of this type, take, for example that of Adad nirari I,^ has a brief 
titulary, then a slightly longer sketch of the campaigns, but the 
greater portion by far is devoted to the narration of his buildings. 
This type also continued until the latest days of the empire, and, 
like the former, is of no value where we have the fuller docu- 
ments. 

When the German excavations were begun at Ashur, the 
earliest capital of the Assyrian empire, it was hoped that the 
scanty data with which we were forced to content ourselves in 
writing the early history would soon be much amplified. In part, 
our expectations have been gratified. We now know the names 
of many new rulers and the number of new inscriptions has been 
enormously increased. But not a single annals inscription from 
this earlier period has been discovered, and it is now becoming 
clear that such documents are not to be expected. Only the so- 
called "Display" inscriptions, and those with the scantiest con- 
tent, have been found, and it is not probable that any will be 
hereafter discovered. 

It was not until the end of the fourteenth century B. C, 
with the reign of Arik den ilu, that we have the appearance of 
actual annalistic inscriptions. That we are at the very begin- 

iBM. 90,978; IV. R. 44f.; G. Smith, Assi/r. Discoveries. 1875^ 
242ff.; Pognon, JA. 1884, 293ff. Peiser, KB. I. 4ff.; Budge-King, 4ff.; 
duplicate Scheil, RT. XV. 138ff.; Jastrow, ZA. X. 35ff; AJSL. XIL 
143ff. 



4 UNIVERSITY OF MISSOURI STUDIES 

ning of annalistic writing is clear, even from the fragmentary 
remains. The work is in annals form, in so far as the events 
of the various years are separated by lines, but it is hardly more 
than a list of places captured and of booty taken, strung together 
by a few formulae.^ 

With this one exception, we do not have a strictly historical 
document nor do we have any source problem worthy of our 
study until the time of Tiglath Pileser I, about 1100 B. C. To 
be sure, we have a good plenty of inscriptions before this time,^ 
and the problems they present are serious enough, but they are 
not of the sort that can be solved by source study. Accordingly, 
we shall begin our detailed study with the inscriptions from this 
reign. Then, after a gap in our knowledge, caused by the tem- 
porary decline of Assyrian power, w^e shall take up the many 
problems presented by the numerous inscriptions of Ashur nasir 
apal (885-860 B. C.) and of his son Shalmaneser III (860-825 
B. C). In the case of the latter, especially, we shall see how a 
proper evaluation of the documents secures a proper apprecia- 
tion of the events in the reign. With these we shall discuss their 
less important successors until the downfall of the dynasty. 
The revival of Assyrian power under Tiglath Pileser IV (745- 
728 B. C.) means a revival of history writing and our problems 
begin again. The Sargonidae, the most important of the various 
Assyrian dynasties, comprising Sargon (722-705 B. C), Sen- 
nacherib (705-686 B. C), Esarhaddon (686-668 B. C), and 
Ashur bani apal (668-626 B. C), furnish us a most embarassing 
wealth of historical material, while the problems, especially as to 
priority of date and as to consequent authority, become most 
complicated. 

Before taking up a more detailed study of these questions, 
it is necessary to secure a general view of the situation we must 

iScheil, OLZ. VII. 216. Now in the Morgan collection, .Johns, Cunei- 
form Inscriptions. 3S. 

2L. Messerschmidt, Keilschrifttexte aus Assur. I. Berlin 1911; Mit- 
thciUmgen der Deutschen Orient Oesellschaft ; cf. D. D. Luckenbill, 
AJSU XXVIII. 153ff. 



ASSYRIAN HISTORIOGRAPHY 5 

face. The types of inscriptions, especially in the later days of the 
empire, are numerous. In addition to the brick and slab inscrip- 
tions, rarely of value in this later period, we have numerous ex- 
amples on a larger scale of the so called "Display" inscriptions. 
They are usually on slabs of stone and are mtended for architect- 
ural adornment. In some cases, we have clay tablets with the 
original drafts prepared for the workmen. Still others are on 
clay prisms or cylinders. These latter do not differ in form from 
many actual annals, but this likeness in form should not blind us 
to the fact that their text is radically different in character. 

AH the display inscriptions are primarily of architectural 
character, whether intended to face the walls of the palace or to 
be deposited as a sort of corner stone under the gates or at the 
corners of the wall. We should not expect their value to be high, 
and indeed they are of but little worth when the corresponding 
annals on which they are based has been preserved. For ex- 
ample, we have four different recensions of a very long display 
inscription, as well as literally scores of minor ones, also of a dis- 
play character, from the later years of Sargon. The minor in- 
scriptions are merely more or less full abstracts of the greater 
and offer absolutely nothing new. The long display inscription 
might be equally well disregarded, had not the edition of the an- 
nals on which it is based come down to us in fragmentary condi- 
tion. We may thus use _th e Dis play inscription to fill gaps in the 
Annals, but it has not the slightest authority when it disagrees 
with its original. 

It is true that for many reigns, even at a fairly late date, the 
display inscriptions are of great value. For the very important 
reign of^Adad nirari (812-785 B. C), it is our only recourse as 
the annals which we may postulate for such a period of develop- 
ment are totally lost. The deliberate destruction of the greater 
portion of the -annals of Tiglath Pileser IV forces us to study the 
display documents in greater detail and the loss of all but a frag- 
ment of the annals of Esarhaddon makes for this period, too, a 
fuller discussion of the display inscriptions than would be other- 
wise necessary. In addition, we may note that there are a few 



O UNIVERSITY OF MISSOURI STUDIES 

inscriptions from otlier reigns, for example, the Xinirud inscrip- 
tion of Sargon, which are seemingly based on an earHcr edition 
of the annals than that which has come down to us and which 
therefore do give us a few new facts. 

Since, then, it is necessary at times to use these display in- 
scriptions, we must frankly recognize their inferior value. We 
must realize that their main purpose was not to give a connected 
history of the reign, but simply to list the various conquests for 
the greater glory of the monarch. Equally serious is it that they 
rarely have a chronological order. Instead, the survey generally 
follows a geographical sweep_from east to west^ That they are 
to be used with caution is obvious. 

Much more fortunate is our position when we have to deal 
with the annalistic inscriptions. We have here a regular chron- 
ology, and if errors, intentional or otherwise, can sometimes 
be found, the relative chronology at least is generally correct. 
The narrative is fuller and interesting details not found in other 
sources are often given. But it would be a great mistake to as- 
sume that the annals are always trustworthy. Earlier historians 
have too generally accepted their statements unless they had defi- 
nite proof of inaccuracy. In the last few years, there has been 
discovered a mass of new material which we may use for the 
criticism of the Sargonide documents. Most valuable are the 
letters, sometimes from the king himself, more often from 
others to the monarch. Some are from the generals in the field, 
others from the governors in the provinces, still others from pal- 
ace officials. All are of course absolutely authentic documents, 
and the light they throw upon the annals is interesting. To 
these we may add the prayers at the oracle of the sun god. com- 
ing from the reigns of Esarhaddon and Ashur bani apal, and they 
show us the break up of the empire as we never should have sus- 
pected from the grandiloquent accounts of the monarchs them- 
selves. Even the business documents occasionally yield us a 
slight help toward criticism. Add to this the references in for- 
eign sources such as Hebrew or Babylonian, and we hardly need 
internal study to convince us that the annals are far from reliable. 



ASSYRIAN HISTORIOGRAPHY 7 

Yet even internal evidence may be utilized. For example, 
when the king is said to have been the same year in two widely 
separated parts of the empire, warring with the natives, it is clear 
that in one of these the deeds of a general have been falsely as- 
cribed to the king, and the suspicion is raised that he may have 
been at home in Assyria all the time.. That there are many such 
false attributions to the king is proved by much other evidence, 
the letters from the generals in comrnand to their ruler/ an occa- 
sional reference to outside authorities, as when the editor of the 
book of Isaiah shows that the famous Ashdod expedition was 
actually led by the Turtanu or prime minister ; or such a docu- 
ment as the dream of Ashur bani apal, which clearly shows that 
he was a frightened degenerate who had not the stamina to take 
his place in the field with the generals whose victories he usurped. 

Again, various versions differ among themselves. To what 
a degree this is true, only those who have made a detailed study 
of the documents can appreciate. Typical examples from Sar- 
gon's Annals were pointed out several years ago.^ The most 
striking of these, the murder of the Armenian king Rusash by 
— the cold blooded Assyrian scribe, — has now been clearly 
proved false by a contemporaneous document emanating from 
Sargon himself. Another good illustration is found in the cool 
taking by Ashur bani apal of bit after bit of the last two Egyp- 
tian campaigns of his father until in the final edition there is 
nothing that he has not claimed for himself. 

The Assyrians, as their business documents show, could be 
exceedingly exact with numbers. But this exactness did not ex- 
tend to their historical inscriptions. We could forgive them for 
giving us in round numbers the total of enemies slain or of 
booty carried off and even a slight exaggeration would be par- 
donable. But what shall we say as to the accuracy of numbers 
in our documents when one edition gives the total slain in a battle 
as 14,000, another as 20,500, the next as 25,000, and the last as 
29,000! Is it surprising that we begin to wonder whether the 

lOlmstead, Western Asia in the Reign of Sargon of Assyria, 1908. 



^J^f^ 






^ 



8 UNIVERSITY OF MISSOURI STUDIES 

victory was only a victory on the clay tablet of the scribe? What 
shall we say when we find that the reviser has transformed a 
booty of 1.235 sheep in his original into a booty of 100,225! 
This last procedure, the addition of a huge round number to the 
fairly small amount of the original, is a common trick of the 
Sargonide scribe, of which many examples may be detected by 
a comparison of Sargon's Display inscription with its original. 
the Annals. So when Sennacherib tells us that he took from 
little Judah no less than 200.150 prisoners, and that in spite of 
the fact that Jerusalem itself was not captured, we may deduct 
the 200.000 as a product of the exuberant fancy of the Assyrian 
scribe and accept the 150 as somewhere near the actual number 
captured and carried off. 

This discussion has led to another problem, that of the rela- 
tive order of the various annals editions. For that there were 
such various editions can be proved for nearly every reign. And 
in nearly every reign it has been the latest and worst edition which 
has regularly been taken by the modern historians as the basis 
for their studies. How prejudicial this may be to a correct view 
of the Assyrian history, the following pages will show. The 
procedure of the Assyrian scribe is regularly the same. As soon 
as the king had won his first important victory, the first edition 
of the annals was issued. With the next great victory, a new 
edition was made out. For the part covered by the earlier edi- 
tion, an abbreviated form of this was incorporated. When the 
scribe reached the period not covered by the earlier document, 
he naturally wrote more fully, as it was more vividly in his mind 
and therefore seemed to him to have a greater importance. Now 
it would seem that all Assyriologists should have long ago recog- 
nized that any one of these editions is of value only when it is 
the most nearly contemporaneous of all those preserved. When 
it is not so contemporaneous, it has absolutely no value zvhen we 
do have the original from which it was derived. Yet it still re- 
mains true that the most accessible editions of these annals are 
those which are the latest and poorest. Many of the earlier and 
more valuable editions have not been republished for many 



ASSYRIAN HISTORIOGRAPHY 9 

years, so that for our most contemporaneous sources we must 
often go to old books, long out of print and difficult to secure, 
while both translation and commentary are hopelessly behind the 
times. Particularly is this the case with the inscriptions of Sen- 
nacherib and Ashur bani apal. The greatest boon to the historian 
of Assyria would be an edition of the Assyrian historical inscrip- 
tions in which would be given only those editions or portions of 
editions which may be considered as contemporaneous and of 
first class value. With such a collection before him, notable 
as much for what it excluded as for what was included, many of 
the most stubborn problems in Assyrian history would cease to be 
problems. 

The historian of Assyria must test his sources before he can 
use them in his history. To do this, he must first of all be able 
to distinguish the primary sources which will reward future 
study from those which are secondary and are based on other and 
more contemporary documents which even now are actually in 
our possession. When these latter are cast aside as of no prac- 
tical value, save perhaps as they show the peculiar mental op- 
erations of the Assyrian editor, we are then ready to test the 
remainder by the various methods known to the historian. The 
second part of this task must be worked out by the historian 
when he studies the actual history in detail. It is the discovery 
of what are the primary sources for the various reigns and of 
the value of the contributions which they make to Assyrian his- 
tory that is to be the subject of the more detailed discussion in 
the following chapters. 



CHAPTER n 

THE BEGINNINGS OF TRUE HISTORY 

(Tiglath Pileser I) 

We shall begin, then, our detailed study of the sources for 
Assyrian history with the data for the reign of Tiglath Pileser I. 
(circa 1100 B. C). Taking up first the Annals, we find that the 
annalistic documents from the reign may be divided into two gen- 
eral groups. One, the Annals proper, is the so called Cylinder, 
in reality written on a number of hexagonal prisms.^ First comes 
the praise of the gods and self praise of the ruler himself. Then 
follow the campaigns, not numbered as in the more developed 
style of later rulers, but separated into six sections, for the six 
years whose events are narrated, by brief glorifications of the 
monarch. Next we have the various hunting exploits of the 
king, and the document ends with an elaborate account of the 
building operations and with threats against the later ruler who 
should destroy the inscription or refuse credit to the king in 
whose honor it was made. 

No relationship has been made out between the fraginents, 
but the four fairly complete prisms fall into two groups, A and 
C, B and D, as regards both the form of writing and the char- 

iPhotographs of B and A, Budge-King, xliii; xlvii; of the Ashur 
fragments, of at least five prisms, Andra, Anu-Adad Tempel, PI. xiii ft. 
I R. 9ff.; Winckler, Sammlung, I. Iff.; Budge-King, 27fE., with variants 
and EM numbers. Lotz, Inschriften Tiglathpilesers I, 1880; Winckler, 
KB. I. 14ff. Rawlinson, Hincks, Talbot, Oppert, JRAS. OS. XVIII. ISOff.; 
Oppert, Histoire des empires de Clialdce et d'Assyrie. 1865, 44f. ; Menant, 
35ff.; Rawlinson, RPi, V. 7£f.; Sayce RP2, I. 92ff.; Muss-Arnolt in Harper, 
llff.; MDOG. 25, 21f.; 28, 22; 29, 40; 47, 33; King, Supplement, 116; 
Andra, Tempel, 32ff. 

(10) 



ASSYRIAN HISTORIOGRAPHY 11 

acter of the text. All date seemingly from the same month of 
the same year, though from separate days. The most fragmen- 
tary of these, D, seems the best, as it has the smallest number 
of unique readings and has also the largest number of omis- 
sions,^ all of which are clearly interpolations in the places 
where they are given. This is especially true of the one^ which 
refers to the Anu-Adad and Ishtar temples, for not only is the 
insertion awkward, we know from the Obelisk^ that the Anu- 
Adad temple was not completed till year five, so that it must be 
an interpolation of that date. In spite of its general resemblance 
to D, especially in its omissions, B is very poorly written and has 
over two hundred unique readings. One of its omissions would 
seriously disarrange the chronology,* others are clearly unwar- 
ranted,^ and one long addition^ further marks its peculiar char- 
acter. Our conclusion must be that it is a poor copy of a good 
original. C is between A and B, agreeing with the latter in a 
strange interpolation"^ and in the omission of the five kings of 
the Muski.8 A is the latest but best preserved, while the char- 
acter of the text warrants us in making this our standard as it 
has but few unique readings and but one improbable omission.^ 
The same account, in slightly different form and seemingly later 
in date^^ is also found in some tablet inscriptions.^^ 

HI. 21b-23a; III. 37b-39a; IV. 36. 

2IV. 36. 

311. 13. 

4IV. 40-42. 

511. 79-81; V. 4; VIII. 29b-33. 

6VII. 17-27; also I. 35; different in VI. 37. 

Till. 2a-c. 

81. 63b. King, Supplement, 116 follows C. 

9VII. 105-8. 

loK. 2815 is dated in the eponomy of Ninib nadin apal, the LAH MA 
GAL E official. He probably is after the rab bi lul official in whose 
year the hexagons are dated. 

iiBudge-King, 125 n.3; K.2815, with different conclusion; 81-2-4, 
220, where reverse different; K.12009; K.13840; 79-7-8, 280; 89-4-26, 28; 
Rm. 573: Winckler. AOF. III. 245. 



12 UNIVERSITY OF MISSOURI STUDIES 

A second annalistic group is that postulated as the original 
of the so called Broken Obelisk. Of documents coming directly 
from Tiglath Pileser himself, the only one that can with any 
probability be assigned to this is the tiny fragment which refers 
to the capture of Babylon.^ But that such a group did exist is 
proved by the extracts from it in the obelisk prepared by a 
descendant of Tiglath Pileser, probably one of his sons, Shamshi 
Adad or Ashur bel kala.^ Only the upper portion, probably less 
than half to judge by the proportions, is preserved, and even this 
is terribly mutilated. Fortunately, the parts best preserved are 
those relating to the years not dealt with in the Annals. The 
first half of the document is devoted to the campaigns of Tiglath 
Pileser, then come his hunting exploits, and only a bit at the 
end is reserved for the building operations of the unknown ruler 
under whom it was erected. Its source seems to have had the 
same relation to the earliest form of the Annals that the Obelisk 
of Shalmaneser III had to the Monolith, that is, it gave the data 
for the earlier part of the reign, that covered by the other source, 
very briefly, only expanding as it reached a period where the 
facts were not represented by any other document. That our 
earlier Annals, or perhaps rather, one of its sources, was a main 
source of our second type, is proved by the coincidences in 
language in the two, in one case no less than twenty signs the 
£'ame,^ not to speak of the hunting expeditions. But this earlier 
Annals was not the only, or at least not the direct source for the 
Obelisk, nor was that source merely a fuller recension of it. 
Data for the first six years, not found in the earlier Annals, are 

IK. 10042; Winckler. AOF. I. 387. 

sphotograph, Budge-King, li; Paterson, Assyr. Sculptures, 63. I R. 
28; III R. 4, 1; Budge-King, 128ff. Lotz, op. cit., 196ff.; Peiser. KB. I. 
122ff.; Talbot. JRAS. OS. XIX. 124ff.; Houghton-Finlay, RPi, XI. 9ff.; 
Oppert, Hist., 132ff.: Hommel, Gesch., 532ff.: Menant, 49fE. Proved to 
Tiglath Pileser, Lotz, op. cit.. 193f.; cf. Budge-King, 131 n. 4, though 
Streck, ZA. XVIII. 187iT., still believes that it belongs to an earlier king. 
Found at Nineveh, though it deals with Ashur constructions. 

3ln year V we have ishtu. ..adi alu Kar garnish sha matu Hatte. ..isu 
elippe pi mashku tahshe. 



ASSYRIAN HISTORIOGRAPHY 13 

given in the Obelisk, ^ while our document also, for the first time 
in Assyrian historical inscriptions, dates the events by the name 
of the eponym for the year, and, still more unusual, by the month 
as well. That the Obelisk may be considered merely a resume 
of this original source is shown by the statement that he con- 
quered other lands and made many wars, but these he did not 
record.2 As they seem to have been given after the hunting feats, 
in the lost lower part of column IV, we may assume that all that 
preceded is taken from that source. Furthermore, we are given 
the other hunting exploits "which my [father] did not record."^ 
The numbers of beasts killed, which the scribe intended especially 
to emphasize, have never, curiously enough, been inscribed in the 
blanks left for their insertion.^ 

Opposed to the Annals proper are the Display inscriptions 
in which chronological considerations and details as to the cam- 
paigns are subordinated to the desire to give a general view of 
the monarch's might. Two have been found in foreign lands, one 
at the source of the Tigris,^ the other near Melazgerd in Ar- 
menia.*^ Drafts for similar inscriptions have been found on clay 
tablets, written for the use of the workmen who were to incise 

lObl. I. 17, reference to Marduk nadin ahe, King of Akkad; II. 1, 
one thousand men of land of . . . ; II. 2, four thousand of them carried 
prisoner to Assyria, the position of which shows that it cannot, with 
Budge-King, 132 n., be referred to Ann. III. 2, the Kashi; II. 12, the 
Mushki (?); II. 13, temple of Anu and Adad. These all precede the 
Carchemish episode. 

20bl. IV. 37. 

30bl. IV. 33. 

4E. g., Obi. IV. 4. 

sDiscovery, J. Taylor, cf. H. Rawlinson, Athenaeum, 1862, II. 811; 
1863, I. 229. Ill R. 4, 6; Schrader, Abh. K. Preuss. Akad.. 1885, I. 
Winckler, Sammlung, I. 30: Budge-King, 127 n. 1. Meissner, Chresto- 
mathie, 6; Abel-Winckler, 5; Menant, 49. Winckler, KB. I. 48f. Dated 
after the Arvad expedition as shown by reference to Great Sea of 
Amurru, and of same date as Melazgerd inscription, Belck, Terh. Berl. 

6From Gonjalu, near Melazgerd, Belck-Lehmann, Terh. Berl. Anthr. 
Ges. 1898, 574. Photograph, Lehmann, Sitzungsber. Berl. Akad., 1900, 
627. Is this one of the "cuneiform inscriptions near Moosh" reported 
to Taylor, Athenaeum, 1863, I. 229? 



14 UNIVERSITY OF MISSOURI STUDIES 

them on stone. Of these, one, which is virtually complete as 
regards number of lines, seems to date from year four as it has 
no reference to later events.^ It would then be our earliest ex- 
tant source. It is also of value in dating the erection of the pal- 
ace whose mention shows that the tablet is complete. That the 
compiler had before him the document used by the Annals in its 
account of the Nairi campaign^ is proved by his writing "from 
Tumme to Daiene" for these are the first and last names in the 
well known list of Nairi states. The order of the tablet is neither 
chronological nor geographical. Another tablet dates from year 
five to which most of its data belong. In the first half, it follows 
the order of Tablet I, and in the remainder follows closely the 
words of its source in the Annals, merely abbreviating.^ Pos- 
sibly in its present form, it may be later than year five^ for a 
third tablet of year ten duplicates this first part.^ Unfortunately, 
this latter gives next to no historical data, but its reference to the 
"Lower Zab" and to the "Temple of Ishtar" may perhaps allow 
us to date to this same tenth year the highly important tablet which 
gives a full account of the campaign in Kirhi and Lulume and 
which also ends with the restoration of the Ishtar temple.'' 
Here too and not with the Annals must be placed the fragment 
with the Arvad episode.'^ 

IS. 1874: K. 2805. Tabl. I of Budge-King, 109ff. Ill R. 5; Winckler, 
Sammlnn p. I. 26f(.; cf. Lotz, op. cif.. 193: Tiele, Gesch., 159 n. 2; Meiss- 
ner, ZA. IX. lOlff. Meissner's restoration of these as parts of one tablet 
in chronological order will not stand in view of the fact that I is com- 
plete in itself while there are variations in the order of Nairi and 
totally different endings. 

2Ann. IV. 71ff. 

3K. 2806 with K. 2804, Tabl. II of Budge-King, 116ff. 

4The badly damaged reverse of K. 2806 has one reference to the 
Euphrates which may be connected with Obi. III. 24, probably of year IX. 

^K. 2804, Tabl. V of Budge-King, 125f. 

6K. 2807; 91-5-9, 196. Ill R. 5, 4; Tablet IV of Budge-King, 121ff. 
Winckler, AOF. III. 246. Hommel, Gesch.. 511f. 

-Scheil, RT. XXII. 157. Restorations, Streck, ZA. XVIII. 186 n. 2. 
First attributed to Tiglath Pileser. Reiser, OLZ. III. 476; Winckler, 
ibid. IV. 296; cf. AOF. III. 247.— Bricks I R. 6, 5: Scheil. op. dt. ?,7; 
Winckler, ffommlung. I. 31; Budge-King, 127. Other inss., King, Supple- 
ment, 453, 488. 



CHAPTER in 

THE DEVELOPMENT OF HISTORICAL WRITING 

(Ashur nasir apal and Shalmaneser III) 

After the death of Tiglath Pileser, there is a period of dark- 
ness. A few bricks and other minor inscriptions give us the 
names of the rulers and possibly a bit of other information, but 
there is not a single inscription which is important enough to 
furnish source problems. It is not until we reach the reign of 
Tukulti Ninib (890-885) that we again have an Annals^ and not 
until the reign of his son Ashur nasir apal (885-860) that we 
have problems of the sources. 

The problem of the sources for the reign of Ashur nasir apal 
may be approached from a somewhat different angle than we 
took for those of Tiglath Pileser. Here we have a single docu- 
ment, the so called Annals, which gives practically all the known 
data of the reign. Earlier writers on the history of Assyria have 
therefore generally contented themselves with references to this 
one document, with, at most, an occasional reference to the 
others. This should not blind us, however, to the fact that the 
problem of the sources is by no means as simple as this. Indeed, 
for far the greater portion of the events given in the Annals, we 
have earlier and better sources. We may therefore best attack 
the problem as to the sources of the reign by working out the 
sources of the Annals. 

Taking up the introduction to the Annals.^ it at once strikes 
us as curious that it consists of a hymn to Ninib, at the entrance 

iScheil, Annales de Tukulti Ninip II, 1909; cf. Winckler, OLZ 
XIII. 112fE. 

21 R. 17ff.; Budge-King, 254ff.; Le Gac, Les Inscriptions (VAssur- 
Nasir-Aplu III. 1907, Iff. Peiser, KB. I. 50ff.; H. Lhotzky, Annalen 

(15) 



16 UNIVERSITY OF MISSOURI STUDIES 

to whose temple these slabs were placed, and not of a general 
invocation to the gods, beginning with Ashur, such as we are 
accustomed to find in other annalistic inscriptions. Further, we 
have other slabs in which this Ninib hymn occurs as a separate 
composition,! and this leads us to assume that it is not the orig- 
inal introduction. This is still further confirmed by the fact that 
we do find such a required invocation in the beginning of the 
MonoHth inscription. Clearly, this is the original invocation. 
The second section of the Annals begins with the praise of the 
monarch, and here too begins the parallelism with the Monolith. 
The last events mentioned in the Monolith date from 880 and 
it is thus far earlier than our present edition of the Annals, which 
contains events from so late a date as 867. To this extent, then, 
the Monolith is a better document. It was not, however, the 
direct source of the Annals, as is shown by certain cases where 
the latter has preserved the better readings of proper names. In- 
deed, we should not over rate the Monolith, for it too is a com- 
pilation like its younger sister, and is by no means free from ob- 
vious mistakes, though in general better than the Annals.^ For 
some portions of this earlier section, we have also separate slabs 
with small portions of the text,^ and these regularly agree with 
the Monolith as against the Annals."* 

Asurnazirpals, 1885. Oppert, Exp<^dition en Mcsopotamie, 1863, I. 311ff.; 
Rodwell, RPi, III. 37ff.; Sayce, RP2, II. 134ff.; Menant, 67ff.; Manuel, 
1880, 335ff. 

iSlabs 27-30, Budge-King, 255 n. — Other invocations are the Bel 
altar at Kalhu, BM. 71, Budge-King 160; Strong, JRAS. 1891. 157; and 
the Ishtar lion BM. 96, II R. 66, 1; S. A. Strong. RP2, IV. 91f.; dupl. 
Budge-King, 206ff. 

2BM. 847. Photograph. Budge-King, Ixix; Paterson, Assyr. Sculp- 
tures, 64. I R. 27; Budge-King. 242ff.; cf. 254fE.; Le Gac, 129fE. Peiser, 
KB. I. 118ff. Menant, 66f.; Talbot, Trans. Roy. Soc. Lit., VII. 189fE.; 
RPi, VII. 15ff. 

3BM. 90830, cf. Budge-King, 255 n.; L. 48f. 

41. 57, transposition; I. 69, the significant omission of shadu; and 
a large number of cases where they agree in spelling as against the 
Annals. 



ASSYRIAN HISTORIOGRAPHY 17 

For the last of these years, 880, we have also the inscription 
from Kirkh/ which contains data for this year alone, and ends 
abruptly with the return from Nairi. This might be expected 
from its location at Tushhan, on the border of that country, and 
we are therefore warranted in assuming that it was set up here 
immediately after the return from the campaign and that in it 
we have a strictly contemporaneous document. Judged by this, 
the Annals, and even the Monolith, do not rank very high. Im- 
portant sections are omitted by each, in fact, they seem to agree 
in these omissions, though in general they agree fairly closely 
with the account set up in the border city. It would seem as if 
the official narrative of the campaign had been prepared at Kirkh, 
immediately after its close, by the scribes who followed the 
army .2 One copy of this became the basis of the Kirkh inscrip- 
tion while another was made at Kalhu and it was from this that 
the Monolith and Annals are derived.^ From this, too, must 
have been derived the slab which gives a fourth witness for this 
section.^ 

With this year, 880, the Monolith fails us. But even if we 
had no other document, the Annals itself would show us that 
the year 880 was an important one in the development of our 
sources. At the end of the account for this year, we have a 
closing paragraph, taken bodily from the Ninib inscription, which 
may thus be assigned to 880. This is further confirmed by the 
manner in which this passage in the Annals abstracts the last 
lines of the Monolith,^ which is repeated almost in its entirety at 

UII R. 6; Budge-King, 222ff.; Le Gac, 137ff. Peiser, KB. I. 92ff. 

2Cf. Johns, Assyr. Deeds and Documents, II. 168. 

3Ann. II. 109, where Mon. has 300 as against 700 of Kir. and Ann., 
shows Ann. did not use Kir. through Mon.; Kir. has 40 as against 50 of 
the others in II. Ill, and 200 for 2000 in II. 115; proper names such as 
Tushha for Tushhan show nearness of Mon. to Kir., but the likeness can 
hardly be considered striking. 

4L. 48f. 

sAnn. II. 125-135a is the same as the Ninib inscription l-23a (BM. 
30; Budge-King, 209ff.), and this in turn is merely a resume of the close 
of the Monolith. 

2 



18 UNIVERSITY OF MISSOURI STUDIES 

the close of the Annals itself. The column thus ends a sepa- 
rate document, whose last line, giving a list of temples erected, 
seems to go back to one recension of the Standard inscription, 
which in its turn goes back to the various separate building in- 
scriptions. 

That the Annals itself existed in several recensions is indi- 
cated by the fact that, while there are no less than at least 
seventeen different duplicates of Column I,^ there are but seven 
of II and five of III ; that there is one of II only^ and one of 
III f and that there is still another, in at least three exemplars. 
in which parts of the Standard and Altar inscriptions are inter- 
polated between the Ninib invocation and the main inscription.* 

The year 880 marks also the removal of the capital from 
Nineveh to Kalhu,^ which indicates that to this year we are to 
attribute the majority of the building inscriptions. But, as they 
are all more or less identical with the closing section of the An- 
nals, we may best discuss them in that place. Continuing with 
the Annals, we now reach a section where it is the only source. 
And just here the Annals is lacking in its most essential feature, 
an exact chronology, no doubt because the dated year was not 
given in the source, though the months are carefully noted ! 
In the last of the years given in this section, probably 876, we 
are to place the various bull and lion inscriptions, which in gen- 
eral agree with this portion of the Annals." One of these bull 
inscriptions, as well as the text of the great altar, adds a good 
bit in regard to the hunting expeditions, which may be dated, so 
far as they can be dated at all, to this year.'^ Here too we must 

iLe Gac, Introd. 

2Le Gac, iii. sjbid. 126f. 

4lbid, ii; 123f. (B). 

^>First mentioned as starting point of an expedition in 879, Ann. 
III. 1. 

6Bulls 76, 77: Lions 809, 841. Budge-King, 189ff.; Le Gac, 181ff. 
Made up of brief attribution to king, then regular building text, then 
duplicates of Ann. III. 84ff. 

TBull 77; Budge-King, 201ff.; Peiser KB. I. 124f.; Altar, L. 43ff.; 
Le Gac, 17 Iff. 



ASSYRIAN HISTORIOGRAPHY 19 

place the Mahir document,^ describing the erection of a temple 
to that deity at Imgnr Bel, as is shown by the specific reference 
to a campaign to the Lebanon for the purpose of securing cedar. 
The years 875-868 seem to have been years of peace, for the only 
reference we can attribute to them is an expedition to the Mehri 
land for beams to erect a temple at Nineveh^ and so to this 
period we must assign the Ishtar bowl inscriptions.-' Finally, 
we have the campaign of 867, the last fixed date in the reign of 
Ashur nasir apal, and the reason for compiling the latest edition 
of the Annals. For this year, and for this alone, this latest edi- 
tion has the value of a strictly contemporaneous document.* 

The last section of the Annals consists of the building ac- 
count, found also in nearly all the other inscriptions, though nat- 
urally here it is in the form it last assumed. It may be seen in 
greater or less fulness in the so called Standard Inscription,^ the 
short account so monotonously repeated on the slabs at Kalhu 
and so familiar to all who have visited any Museum where As- 
syrian antiquities are preserved. There seem to be two recen- 

iV R. 69f.; Budge-King, TSBA. VII. 59ff.; Budge-King, 167ff. S. A. 
Strong, RP2, IV. 83ff.; Harper, 29ff. 

2Ann. III. 91f. 

3III R. 3, 10; Budge-King, 158ff.; S. A. Strong, RP2, II. 95. 

4Ann. III. 92ff. 

5L. Iff.; Schrader, Inschrift Asur-nasir-abals ; Taltot, Proc. Soc. 
Antiquaries of Scotland, VI. 198ff.; Meissner, Chestomathie, 7f. ; Abel- 
Winckler, 6. RPi, VII. llff.; Ward, Proc. Amer. Oriental Soc, X. xcix; 
Budge-King, 212ff. ; Le Gac, 153ff. The number of slabs containing this 
inscription whicli may be found in the various Museums of Europe and 
America is simply amazing. No full collection or collation of these has 
ever been made. Many are still exposed to the destructive effects of the 
atmosphere at Nimrud and are rapidly being ruined. Squeezes of these 
were taken by the Cornell Expedition. Others at Ashur, MDOG., xxi. 
52; KTA. 25. Several are in the newly opened section of the Constan- 
tinople Museum, cf. Bezold, Ztf. /. KeilschriftforscJiung, I. 269. An un- 
known number is in the British Museum, and were utilized by Budge- 
King, 1. c. Streck. ZA. XIX. 258, lists those published from European 
Museums. These are Edinburgh, Talbot 1. c; Copenhagen, Knudtzon, 
ZA. XII. 256; St. Petersburg, Jeremias, ZA. I. 49; Bucharest, D. H. 
Miiller, Wiener Ztf. J. Eunde d. Morgenlandes, XIII. 169ff.; Dresden, 



20 UNIVERSITY OF MISSOURI STUDIES 

sions, a longer and a shorter,^ and some, to judge from the 
variations in the references, are much later than 880. The 
same inscription essentially is also found as the ending of the 
Ishtar, Mahir, Calah Palace,^ Calah wall,^ Bulls, and Ninib in- 
scriptions."* Variants are few, but are not without value in fix- 
ing the relative dates of the various recensions. For example, 
some of the Standard inscriptions, as well as the Ishtar and 
jNIahir ones, insert a reference to "Mount Lebanon and the Great 
Sea" which would place them after 876, and this is confirmed by 
the reference to Liburna of Patina which occurs in the Annals 
and the Calah wall inscription. Of course, this gives only the 

Jeremias, I. c; Zurich, Bezold, Literatur, 71; Cannes, Le Gac, ZA. IX. 
390; Lyons, Ley, RT. XVII. 55; Rome, O. Marucchi, Museo Egizio 
Vaticano, 334; Bezold, ZA. II. 229. In addition, there are, according to 
Budge-King, 7. c. copies at Paris, Berlin, Munich, the Hague, etc. For 
the Berlin inscriptions, cf. Verzeichnis der vorderasiatischen Alter- 
turner, 92ff. ; 101. No less than 59 are known to have been or to be in 
America. The majority have been listed by Ward, op. cit., xxxv, and 
Merrill, ibid., xci. ff. : cf. BibUotheca Sacra, xxxii. 320fl. Twelve in the 
possession of the New York Historical Society have not been on exhibi- 
tion since the society moved into its new quarters, and are completely 
inaccessible, the statements in the guide books to the contrary notwith- 
standing. The Andover slab is published by Merrill, op. cit. Ixxiii, and 
the one from Amherst by Ward, I. c. These were presented by Rawlin- 
son and Layard to missionaries, and by them to the institutions named, 
as were the following: Yale University; Union College, Schenectady; 
Williams College; Dartmouth College; Middlebury College; Bowdoin 
College; Auburn (N. Y.) Theological Seminary; Connecticut Historical 
society at Hartford; Meriden (Conn.) Public Library; Theological 
Seminary of Virginia; Mercantile Library of St. Louis. An inscribed 
relief to which my attention has been called by Professor Allan Mar- 
quand, has been presented by Mr. Garrett to Princeton University. Three 
similar slabs, loaned by the late Mr. J. P. Morgan, are in the Metropolitan 
Museum in New York City. — In this place we may also note the brick 
inscriptions in America, listed by Merrill, /. c. as well as the statute 
inscription. III R. 4, 8; Menant, 65; Schrader, Keilinschriften und das 
Alte Testament,'^ 184. 

iLe Gac, xvii. 

2Budge-King, 173ff.; Le Gac, 188ff. 

3Budge-King, ITTff. 

4Budge-King, 209ff. 



ASSYRIAN HISTORIOGRAPHY 21 

upper limit, for it would be dangerous to suggest a lower one in 
the case of documents which copy so servilely. Some of the 
Standard inscriptions, as well as the Bulls, have a reference to 
Urartu, of great importance as the first in any literature to the 
country which was soon to become the worthy rival of Assyria. 
Absence of such reference in the regular Annals is pretty con- 
clusive evidence that there were no warlike relations, so that 
these too are to be dated after 876. With this is to be compared 
the addition telHng of the conquest of Nairi, found in the Ishtar, 
Mahir, and Calah Palace inscriptions, and which would seem to 
refer to the same period. The Suhi, Laqe, and Sirqu reference, 
through its omission in the Monolith, is also of value as adding 
proof that that inscription dates to 880.^ 

Much the same situation as regards the sources is found in 
the reign of his son Shalmaneser III (860-825). Aside from a 
few minor inscriptions, our main source is again the official ac- 
count which has come down to us in several recensions of dif- 
ferent date. The process by which these recensions were made 
is always the same. The next earlier edition was taken as a 
basis, and from this were extracted, generally in the exact words 
of the original, such facts as seemed of value to the compiler. 
When the end of this original was reached, and it was necessary 
for the editor to construct his own narrative, the recital becomes 
fuller, and, needless to say, becomes also a better source. If, 
then, we have the original from which the earliest portion of a 
certain document was copied or abstracted, we must entirely 
cast aside the copy in favor of the contemporary writing. This 
would appear self evident, but failure to observe this distinction 
has led to more than one error in the history of the reign.^ 

iMinor inscriptions, L. 83f.; G. Smith, Disc, 76; Budge-is.ing, 155ff.". 
Le Gac, 172; the very fragmentary Obelisk, Le Gac, 207ff.; KTA. 25; 
MDOG. 20, 21ff.; 21, 15ff. King, Supplement, no. 192, 470, 1805. Hommel. 
Zwei Jagdinschriften, 1879, with photographs; Andra, Tempel, 86ff. 

2The majority of the inscriptions for the reign were first given in 
Layard, Inscriptions, and in the Rawlinson publication, cf. for first work- 
ing over, Rawlinson, JRAS. OS. XII. 431ff. The edition of Amiaud- 
Scheil, Les inscriptions d'e Sahnanasar II, 1890, though without cunei- 



22 UNIVERSITY OF MISSOURI STUDIES 

Each of these editions ends with the account of some im- 
portant campaign, the need of writing up which was the reason 
for the collection of the events of previous years which were not 
in themselves worthy of special commemoration. The first of 
these is the one which ends with the famous battle of Qarqara 
in 854. This has come down to us in a monumental copy which 
was set up at Kirkh, the ancient Tushhan, and which has been 
named the Monolith inscription.^ For the events of 860-854, 
then, we need go no further than this, for it is strictly contempo- 
raneous with the events it describes. No actual errors can be 
pointed out in it, a seeming distortion of the chronology being 
due simply to the desire of the scribe to indicate the unity of 
two campaigns, carried out in different years, but against the 
same country .^ How moderate are its numbers is shown by com- 
paring its 14,000 killed at Qarqara with the 20,500 of the Obe- 
lisk, the 25,000 of the Bulls, and the 29,000 of the recently dis- 
covered statue from Ashur. As we shall see below, it is correct 
in giving no campaign for 855, though the Bulls inscription, writ- 
ten a generation later, has not hesitated to fill the gap. This is 
the only edition which seems to be entirely original and a com- 
parison with those which are in large part compilations is favor- 
able to it in every way. In fact, the oft repeated reproach as to 
the catalogue nature of the Shalmaneser writings is due to the 
taking of the Obelisk as a fair sample, whereas it stands at the 
other extreme, that of a document almost entirely made up by 
abridgement of other documents, and so can hardly be expected 
to retain much of the literary flavor of its originals. The INIono- 

form text, is still valuable on account of its arrangement by years, as 
well as of its full notes, cf. also Winckler-Peiser, KB. I. 128ff. The one 
edition which is up to date is N. Rasmussen, Sahnanasser den IPs Ind- 
skriften, 1907, though the same may be said of the selections in Rogers, 
293ff. 

iIII R 7f; Rasmussen, Iff.; 2ff. Photograph. Rogers, 537; Hist., 
op. 226. Amiaud-Scheil, passim: Reiser, KB. I. 150ff. Menant, 105ff.; 
Sayce, RPi, III. 83ff.; Scheil, RP2, IV. 55ff.; Craig, Hehraica, III. 201ff.; 
Harper, 33ff.; cf. Jastrow, AJSL. IV. 244ff. 

211. 66. 



ASSYRIAN HISTORIOGRAPHY 23 

lith, on the other hand, free from the necessity of abridging, will 
hold its own in literary value with the other historical writings 
of the Assyrians. 

The next edition was prepared in 851, at the conclusion of 
the Babylonian expedition. The document as a whole is lost, but 
we have excerpts in the Balawat inscription. ^ For the years 859, 
857, and 856, the excerpts are very brief, but fortunately this is 
of no importance as we have their originals in the Monolith. 
No mention is made of the years following until 852-851 which 
are described so fully that we may believe we have here the 
actual words of the document. It is interesting to notice that 
there is no particular connection between the reliefs on the 
famous bronzes- and the inscription which accompanies them. 
The latter ends in 851, the pictures go on to 849. The more con- 
spicious pictures were brought up to date, but, for the inscription 
which few would read, a few extracts, borrowed from the edi- 
tion of two years previous, sufficed. Incidentally, it shows us 
that no new edition had been made in those two years. For the 
years before 853, the practical loss of this edition need trouble 
us little as it seems merely to have copied the original of the 
Monolith. That it might have had some slight value in restoring 
the text of that lost original seems indicated by a hint of a fuller 
text in one place-^ and a more moderate number of enemies 
slaughtered in another.'* For the events of 853, as given in this 

iPinches, PSBA. VII. 89fE.; The Bronze Ornaments of the Palace 
Gates of Balawat, 1880; Rasmussen, Xlff. ; Amiaud-Scheil, passim; 
Delitzsch, Beitr. z. Assyr., VI. 133ff.; Winckler KB. I. 134ff. Scheil, 
RP2, IV. 74ff. 

sPinches, Bronze Ornaments, a magnificent publication. A cheaper 
edition of the reliefs, with valuable analysis of and comments on the 
sculptures, Billerbeck; Beitr. z. Assyr. VI. Iff. Additional reliefs owned 
by G. Schlumberger, Lenormant, Gazette Arch., 1878 pi. 22ff. and p. 
119ff. Still others, de Clerq, Catalogue, II 183fC., quoted Billerbeck, 2. 
I have not yet seen King, Bronze Reliefs from the Gates of Shalmaneser, 
1915. 

311. 6f. 

4Balawat kills but 300 while Monolith slaughters 3400. 



24 UNIVERSITY OF MISSOURI STUDIES 

edition, we have only the abstract of it in the Bulls in- 
scription.^ 

The year 845, the year of the expedition to the sources of 
the Tigris, seems to mark the end of a third period, commemo- 
rated by a third edition, extracts from which are given in the in- 
scriptions on the Bulls.- That it actually began with the year 
850 is shown by the use of a new system of dating, by the king's 
year and the number of the Euphrates crossing. Comparison 
with passages preserved in the Balawat extracts shows that the 
work of excerpting has been badly done by the editor of the third 
edition. The capture of Lahiru is placed in the wrong year,"^ 
the graphical error of Ukani for Amukkani shows it derived from 
the Balawat edition, while variations between the two copies of 
the bull inscription indicate that we cannot be sure of the exact 
words of the original.* And we can also point to deliberate 
falsification in the insertion of an expedition to Kashiari against 
Anhitti of Shupria, when the older edition, the Monolith, knew 
of no expedition for the year 855. It has already been shown 
elsewhere that this is closely connected with the attempt of the 
turtanu (prime minister) Dan Ashur to date his accession to 
power to 856 instead of 854, and to hide the fact of the palace 
revolution which seems to have marked the year 855.-^ 

From various hints, it is possible to prove that a fourth edi- 
tion was prepared in 837, the end of the wars with Tabal. The 
most striking evidence for this is the fact that, after this year, the 
Obelisk suddenly becomes much fuller, a clear proof that the 
author knew that he was now dealing with events not previously 
written up. We may see, then, in the Obelisk account from 844 
to 837 an abstract of the lost edition of 837. But we are not con- 

iBull 75ff. 

zDiscovery, Layard, NR. I. 59. L. 12£E.; 46f.; Rasmussen, XVff.; 
42ff. Amiaud-Scheil, passim: Delitzsch, op. cit.. 144ff.; Menant, llSff. 

3Bull 79; cf. Balawat IV. 6. 

^Variants in Amiaud-Scheil, passim. The most striking is the 
different text with which they end, cf. Amiaud-Scheil. 58 n. 1. 

5Cf. below under the Obelisk, and, for fuller discussion, Olmstead, 
Jour. Amer. Or. Soc. XXXIV. 346f. 



ASSYRIAN HISTORIOGRAPHY 25 

fined to this. One actual fragment of this edition is the frag- 
ment which deals with the events of 842 and is so well known 
because of its reference to Jehu.^ The first half of this is also 
intercalated after the introduction to one of the Bull inscrip- 
tions, and before year four, thus showing that it was inserted to 
bring the edition of 845 up to date.^ Based on this edition, 
though only in very brief abstract, seems also the so called throne 
inscription from Ashur, whose references to Damascus, Que, 
Tabal, and Melidi form a group which can best be correlated 
with the events of the years 839, 840, 838, and 837, respectively .^ 
Another Ashur inscription on a royal statute gives selections 
from the events of the reign, up to 835, but its main source is 
evidently the same.^ 

But the strongest proof of the existence of this edition is to 
be found in the two fragments of clay tablets which are not, like 
all the preceding, epigraphical copies of the originals, but form 
part of the original itself.^ These two bits are written in the 
cursive style, and, though their discoverer believed them to be- 
long to separate documents, the fact that one so closely supple- 
ments the other, and that they have the same common relation 
to the other editions, justifies us in assuming that they really 
do belong together. At first sight, it might be argued that they 
are to be restored from the text of the Obelisk, with which they 
often agree verbally. Closer inspection shows, however, that 
they contain matter which is not found in that monument, and 
that therefore they belong to an earlier and fuller edition, yet the 

iIII R. 5, 6; Rasmussen, XXI; 56; Delitzsch, Assyr. LesestUcke'^, 
51f. AmiaudnScheil, 58; Winckler, KB. I. 140; Ungnad, I. 112; Rogers, 
303f. 

2L. 121; Rasmussen, XIX; 53. 

sDiscovery, Layard, NR. II. 46ff.; cf. G. Smith, TSBA. I. 77. L. 
76f.; Craig, He&raica, II 140ff.; Rasmussen, XXXVIII; 84ff.; Amiaud- 
Scheil, 74ff.; Delitzsch, Beitr. z. Assyr., VI. 152f.; cf. Jastrow, Hebraica, 
V. 230ff. 

4Andra, MDOG. 21, 20ff.; 39ff.; Delitzsch, ibid. 52; KTA. 30; Lang- 
don, Expository Times, XXIII, 69; Rogers, 298f.; 529. 

sBoissier, RT. XXV. 82ff. 



26 UNIVERSITY OF MISSOURI STUDIES 

resemblance to the Obelisk is so close that they cannot be much 
earlier. On the other hand, the Bulls inscription can be com- 
pared for the events of 854-852 and this has all that our tablets 
have, plus a good bit more. They therefore belong between 
these two editions, and the only time we can place them is 837. 
Since the clay tablets so fully abstract the Bulls inscription 
wherever the latter is available for comparison, we may assume 
that in 857-855 they give the minimum of that inscription. Thus 
we have the editions of 845, of 837, and of 829, in a common line 
of descent. Although for 857-856, there are numerous verbal 
coincidences with the Balawat excerpts, it must be noted that not 
all the plus of our tablets appears in that document, and we can 
only assume a common source, a conclusion which well agrees 
with our characterization of the Balawat inscription as a series 
of mere extracts. That this common source was also the source 
of the Monolith seems proved by a certain similarity of phrase- 
ology as well as by the reference to Tiglath Pileser in connection 
with Pitru, but this similarity is not great enough fully to re- 
store our plus passages. Unfortunately for the student of his- 
tory, our tablets do not add any new facts, for, in the parts 
preserved, we already had the earlier representatives of the 
original sources from which the edition was derived. It does, 
however, throw a most interesting light on the composition and 
development of these sources. 

Last and least valuable of all is the Obelisk.^ Because of 
its most interesting sculptures and because it gives a summary 
of almost the entire reign, it has either been given the place of 
honor, or a place second to the Monolith alone. The current 

iDiscovery at Kalhu, Layard, NR. II. 282. Layard, Monuments of 
Nineveh. I. 53ff.; L. 87ff.; Abel-Winckler, 7f; Rasmussen, XXXIIIff.; 80ff. 
Amiaud-Scheil. passim; Winckler. KB. I. 128ff Oppert, Exvcd. I. 342 
Hist. 108ff.: Menant, 97ff.; Sayce, RPi, V. 29ff.; Scheil, RP2, IV. 38 
Jastrow, Hcbraica, V. 230. Mengedoht, Bab. Or. Rec, VIII, lllff.; 141£E. 
169ff. Photographs and drawings too frequent for notice. Casts are also 
common, e. g., in America, Metropolitan Museum, N. Y. City; University 
of Pennsylvania; Haskell Museum, University of Chicago; Boston Mu- 
seum of Fine Arts. 



ASSYRIAN HISTORIOGRAPHY 27 

view is given by one of our most prominent Assyriologists as 
follows : "The first rank must be ascribed to the Black Obelisk, 
and for the reason that it covers a greater period of Shalman- 
eser's reign than any other. ... It is clear then, that for a study 
of the reign of Shalmaneser II the black obelisk must form 
the starting point, and that, in direct connection with it, the 
other inscriptions may best be studied, grouping themselves 
around it as so many additional fragmentary manuscripts would 
around the more complete one which we hit upon, for a funda- 
mental text."i 

This view might be accepted were the problem one of the 
"lower criticism". Unfortunately, it is clearly one for the 
"higher" and accordingly we should quote the Black Obelisk only 
when an earlier edition has not been preserved. There is no 
single point where, in comparison with an earlier one, there is 
reason to believe that it has the correct text, in fact, it is, as 
might be expected in the case of a show inscription, filled with 
mistakes, many of which were later corrected, while in one case 
the engraver has been forced to erase entire lines.^ Its date is 
829, a whole generation later than the facts first related, and it 
can be shown that it is a formal apology for the turtanu (prime 
minister), Dan Ashur, glorifies him at the expense of his mon- 
arch, and attempts to conceal the palace revolution which marked 
his coming into power by changing the date of his eponomy from 
854 to 856 and by filling in the year 855 with another event. 
Nor is it without bearing in this connection that it was prepared 
in 829, the very year in which the revolt of Ashur dan apal broke 
out as a protest against the control of his father by the too 
powerful turtanu.^ As these last years of the reign were years 
of revolt, there is no reason for believing that there was another 
edition prepared, and the narrative of this revolt in the Annals 
of his son Shamshi Adad points in the same direction. 

iJastrow, I. c. 

2Cf. the textual commentary in Amiaud-Scheil, passim, and es- 
pecially 65 n. 6. 

3Cf. Olmstead, Jour. Amer. Or. Soc, I. c. 



28 UNIVERSITY OF MISSOURI STUDIES 

Of documents which do not belong to this connected series, 
the most important is the recently discovered lion inscription 
from Til Barsip. Aside from its value in identifying the site of 
that important city and an extra detail or two, its importance 
is not great, as it is the usual type of display inscription.^ The 
Tigris Tunnel inscription also has its main importance from the 
locality in which it was found.^ Other brief inscriptions add a 
bit as to the building operations, which, curiously enough, are 
neglected in the official annals series.'^ 

iR. C. Thompson, PSBA. XXXIV. 66ff.; cf. Hogarth, Accidents of 
a7i Antiquary's Life. op. 175. 

^Scheil, RT. XXII. 38. 

3L. 77f.; Amiaud-Scheil, 78; Rasmussen, XLI; 88f. Layard, NR. 
II. 46; I. 281. Bricks in America, Merrill, Proc. Amer. Or. Soc. X. c; 
Bibl. Sacra. XXXII. 337ff.; Streck, Ztf. Deutsch. Morg. Gesell.. 1908, 758; 
Scheil, RT. XXVI. 35ff.; Pinches, PSBA. XXXII. 49f., of year I; KTA. 
26ff.; 77; MDOG. 21, 20f.; 22, 29ff.; 22, 77; 28, 24f.; 31, 15; 32, 15ff.; 
36, 16ff.; 48, 27; Andra, Tempel, 41ff.; Taf. XX. XXIIf. 



CHAPTER IV 

SHAMSHI ADAD AND THE SYNCHRONISTIC HISTORY 

The main source for the reign of Shamshi Adad (825-812) 
is the official Annals which exists in two recensions. One, writ- 
ten in archaistic characters, from the south east palace at Kalhu, 
has long been known. After the usual introduction, it deals 
briefly with the revolt of Ashur dan apal. Xo attempt is made 
to differentiate the part which deals with his father's reign from 
that of his own, and the single paragraph which is devoted to 
it gives us no real idea of its importance or of its duration. Then 
follow four expeditions, the first two given very briefly, the last 
rather fully. As the years of the reign are not indicated, there 
is considerable difficulty in obtaining a satisfactory chronolog)'.^ 
The other carries the record two years further, but has not yet 
been published.^ 

The long list of expeditions which the Assyrian Chronicle 
attributes to the reign of Adad nirari (812-783) indicates that 
he must have composed Annals, but they have not as yet been 
discovered. Of extant inscriptions, the earliest is probably that 
on the statue base of Sammuramat (Semiramis), in which she 
is placed before her son and emphasis is laid on the fact that she 
is the widow of Shamshi Adad rather than that she is the mother 
of the reigning monarch.^ Next in time comes the inscription 
on the famous Nabu statue in which x\dad nirari is placed first, 

iIR. 29ff. Scheil, Inscription Assyr. Archaigue de Samsi Ramman 
IT, 1889. Abel, KJB. I. 174ff. Oppert, Hist., 122ff.; Menant, 119£E.; 
Sayce, RPi, I. llff. ; Harper, 45ff. For errors in writing cf. Scheil, VI; 
for use of rare words, ibid. VII. 

2MD0G. 28, 3 If. Through the courtesy of Dr. Andra, I was per- 
mitted to see this in the excavation house at Ashur in 1908. — Cf. also 
the palace brick, Scheil, RT. XXII. 37. 

3MD0G. 40, 24ff. 42, 34ff. 

(29) 



30 UNIVERSITY OF MISSOURI STUDIES 

but with Samnuiramat at his side, and which accordingly marks 
the decline of the queen mother's power. ^ Near the end of his 
reign must be placed the two Kalhu inscriptions in which Sam- 
muramat is not mentioned. One refers to the conquests from 
the sea of the rising sun to the sea of the setting sun, a statement 
which would be possible only after the conquest of Kis in 786. 
This is the document which throws a vivid light on the early 
history of Assyria, but the remainder is lost^ and a duplicate adds 
nothing new.^ The other Kalhu inscription adds considerable 
material, but in a condensed form which makes it most difficult 
to locate the facts in time. The historical portion is divided into 
three sections which seem roughly to correspond with the chrono- 
logical order. First comes a list of the peoples conquered on the 
eastern frontier, arranged geographically from south to north. 
As but two of these names are listed in the Assyrian Chronicle, 
and as each occurs several times, it is impossible to locate them 
exactly in time. The second section deals in considerable detail 
with an expedition against Damascus but the Chronicle does not 
list one even against central Syria. The fulness of this account 
shows that it took place not far from the subjugation of Kaldi 
land, the narrative of which ends the document and shows it to 
have been written not far from 786, its date in the Chronicle.* 
For the remaining reigns of the dynasty, we have only the 
data in the Assyrian Chronicle. No annals or in fact any other 
inscription has come down to us, and, so far at least as the 

iRawlinson, Monarchies, II. 118 n. 7; Photograph, Rogers, 511; 
Religion, op. 86; I. R. 35, 2; Abel-Winckler, 14; Abel, KB. I. 192f.; 
Rogers, 307f.; Winckler, Textbtich^, 27f.; Meissner, Chrestomathie, 10; 
Menant, 127f. 

2Layard, NR. II. 20. L. 70; I. R. 35, 3; Delitzsch, Lesestiicke^, 99; 
Abel-Winckler, 13. Abel, KB. I. 188ff. Sayce, RPi, I. 3ff.; S. A. Strong, 
RP2, IV. 88f.; Harper, 50f. 

3L. 70. 

4Rawlinson, Athenaeum, 1856, 174; I R. 35, 1; Winckler, Text- 
buch^, 26f. Abel, KB. I. 190ff.; Ungnad, I. 112f.; Rogers, 306f. Tal- 
bot, JRAS. XIX. 182ff.; Harper, 51f.; Meissner, Chrestomathie. 9; 
Menant, 126f.— Nineveh brick, I R. 35, 4. Abel, KB. I. 188f. Ashur in- 
scriptions, KTA. 35f.; MDOG. 22, 19; 26, 62. 



ASSYRIAN HISTORIOGRAPHY 31 

annals are concerned, there is little likelihood of their discovery, 
as there is no reason to believe that any were composed in this 
period of complete decline. But, curiously enough, from this 
very period comes the document which throws the most light 
on the earliest period of Assyrian expansion, the so called Syn- 
chronistic history.^ Adad nirari is the last ruler mentioned, but 
the fact that he is named in the third person shows that it was 
compiled not earlier than the reign of his successor Shalman- 
eser IV. 

Our present copy is a tablet from the library of a later king, 
seemingly Ashur bani apal.^ In form, it marks an advance over 
any historical document we have thus far studied, for it is an 
actual history for many centuries of the relations between As- 
syria and Babylonia. But it is as dry as possible, for only the 
barest facts are given, with none of the mass of picturesque de- 
tails which we have learned to expect in the annals of the indi- 
vidual kings. Nevertheless, its advance over preceding docu- 
ments should not be over estimated. Its emphasis on treaties 
and boundaries has led to the idea that it was compiled from the 
archives as a sort of diplomatic piece justificative in a contro- 
versy with Babylonia over the possession of a definite territory.^ 
Its true character, however, is clearly brought out in its closing 
words "A succeeding prince whom they shall establish in the 
land of Akkad, victory and conquest may he write down, and on 
this inscribed stone (naru), eternal and not to be forgotten, may 
he [add it]. Whoever takes it, may he listen to all that is writ- 
ten, the majesty of the land of Ashur may he worship contin- 
ually. As for Shumer and Akkad, their sins may he expose to 
all the regions of the world. "^ 

III R. 65, 1; III R. 4, 3; Winckler, Untersuch., 148ff.; CT. XXXVI. 
38ff.; cf. the introduction of Budge-King; King, Tulculti Ninih. Peiser- 
Winclder, KB. I. 194ff.; G. Smith, Disc. 250f.; Sayce, TSBA. II. 119ff.; 
RPi, III. 29fE.; RP2, IV. 24ff.; Barta in Harper, 196; cf. Winckler, AOF. 
I. 114ff.; Belck, Beitr. Geop. Gesch., I. 5ff. 

2Maspero, Hist., II. 595, dates its composition to this reign. 

sPeiser-Winckler, KB. I. 194 n. 1. 

4IV. 22ff. 



32 UNIVERSITY OF MISSOURI STUDIES 

Obviously, then, this tablet of clay is only a copy of an 
earlier naru or memorial inscription on stone, and we should ex- 
pect it to be only the usual display inscription. This is still fur- 
ther proved by the introduction, mutilated as it is, ". . .to the god 
Ashur...his prayer. . .before his face I speak .... eternally a 
[tablet] with the mention. .. .the majesty and victory [which 
the kings of Ashur mad]e, they conquered all, [the march] 

of former [expedi]tions, who conquered [their booty to 

their lands they br]ought. .." Clearly, this is the language of 
a display inscription and not of a diplomatic piece justificative. 
So we can consider our document not even a history in the true 
sense of the word, merely an inscription erected to the glory of 
Ashur and of his people, but with the "sins of Shumer and Ak- 
kad," in other words, with the wars of the Babylonians against 
"the land"! and w^ith the sinful destruction of Assyrian property 
they caused, also in mind. When we take this view, we are no 
longer troubled by the numerous mistakes, even to the order of 
the kings, which so greatly reduce the value of the document 
where its testimony is most needed.^ We can understand such 
"mistakes" in a display inscription, exposed to view in a place 
where it would not be safe for an individual to point out the 
truth. But that it could have been used as a piece justificative, 
with all its errors, when the Babylonians could at once have re- 
futed it, is incredible. 

The accession of Tiglath Pileser IV (745-728) marks a 
return to warfare, and the consequent prosperity is reflected in 
an increase of the sources both in quantity and in quality.^ 
Tiglath Pileser prepared for the walls of his palace a series of 
annals, in three recensions, marked by the number of lines to 
the slab, seven, twelve, or sixteen, and seemingly by little else. 
Originally they adorned the walls of the central palace at Kalhu, 

iCf. Belck, Beit7: Geog. Gesch. I. 5ff. — The double mention of Ashur 
bel kala and Shalmaneser points to double sources, one the original of 
BM. 27859, Peiser, OLZ. XI. 141. 

2Cf. Winckler, AOF. I. 109£f. 

3For inscriptions of reign, cf. Rost, Keilschrifttexte Tiglat-Pilesers 
III; cf. also Anspacher, Tiglath Pileser, Iff. 



ASSYRIAN HISTORIOGRAPHY 33 

but Esarhaddon, a later king of another dynasty, defaced many 
of the slabs and built them into his south west palace. Thus, 
even with the three different recensions, a large part of the An- 
nals has been lost forever. For years, the great problem of the 
reign of Tiglath Pileser was the proper chronological arrange- 
ment of this inscription. Thanks to the aid of the Assyrian 
Chronicle, it is now fairly fixed, though with serious gaps. Once 
they are arranged, little further criticism is needed, for they are 
the usual type, rather dry and uninteresting to judge from the ex- 
tant fragments.* Perhaps separate notice should be given to the 
sculptured slabs in Zurich with selections from the Annals.^ 

Next to the Annals comes the clay tablet from Kalhu, from 
which, if we are to judge by the proportions, less than a half has 
survived.-^ Thus, owing to the method used by the Assyrians in 
turning the tablet for writing, only the first and last parts are 
preserved. Unfortunately, the greater part of what is preserved 
is taken up with an elaborate introduction and conclusion which 
we would gladly exchange for more strictly historical data. The 
other contents are, first an elaborate account of the wars in Baby- 
lonia, next of the wars on the Elamite frontier, a brief paragraph 
on Ulluba and Kirhu, and then the beginning of the war with 
Urartu. Each of these paragraphs is marked off by a line across 
the tablet. Thus far, it is clear, we have a geographical order for 

iDetailed bibliography of the fragments, Anspacher, Tiglath Pileser, 
3ff.; Discovery, Layard, NR. II. 300. L. 19ff.; Ill R. 9f. Rost, de in- 
scriptione Tiglat-Pileser III quae vacatur Annalium. 1892; Rost, Iff.; 
2ff.; "Winckler, Textbuchs, 28ff. Ungnad I. llBff.; Rogers, 313ff.; Schrader 
KB. II. 24ff.; Rodwell, RPi, V. 45ff.; Menant, 144ff. For discussion of 
arrangements of fragments, cf. G. Smith, Ztf. f. Aegyptologie, 1869, 9fE.; 
Disc, 266; Schrader, Keilschrift und Geschichtsforschung, 395ff. ; Abh. 
Berl. Akad., 1880; Tiele, Gesch., 224; Hommel, Gesch., 648ff. 

2Boissier, PSBA. I have not seen his Notice sur quelque Monuments 
Assyr. a V university de Zilrich, 1912. 

sUsually called the Nimrud inscription, a cause of confusion. K. 
3751. Photograph of obverse, but upside down, Rogers, 541; History, op. 
267. II R. 67; Rost, XXXVff; 54ff. Schrader, KB. II. 8ff.; Erneberg, 
JA. VII. Ser. VI. 441ff.; Menant, 140ff.; Smith, Disc,, 256ff.; Strong, 
RP2, V. 115fE.; J. M. P. Smith, in Harper, 52ff.; Rogers, 322. 

3 



34 UNIVERSITY OF MISSOURI STUDIES 

the paragraphs. After the break, we have an account of the Arab 
tribes on the border of Egypt. It is therefore clear that the or- 
der was continued in the break which must have contained the 
most of the Urartu account and whatever was said about Syria. 
The fulness with which the extant portion chronicles the Baby- 
lonian affairs makes it probable that the part now lost in the 
break dealt with Armenian and Syrian relations with equal ful- 
ness. The next paragraph seems to be a sort of summary of the 
various western rulers who had paid tribute, and the length of 
this list is another proof of the large amount lost. The very 
brief Tabal and Tyre paragraphs, out of the regular geographical 
order, are obvious postscripts and this dates them to year XVII 
(729), unless we are to assume that the scribe did not have them 
in mind when he wrote the reference to that year in the introduc- 
tion. That they really did date to the next year, 728, is indicated 
by the fact that the Assyrian Chronicle seems to have had a 
Tyre expedition in that year.^ If so, then our inscription must 
date from the last months of Tiglath Pileser's reign. Though 
written on clay, it is clearly a draft from which to engrave a dis- 
play inscription on stone as it begins "Palace of Tiglath Pileser." 
The identity of certain passages- with the Nimrud slab shows 
close connection, but naturally the much fuller recital of the tab- 
let is not derived from it. We have also a duplicate fragment 
from the Nabu temple at Kalhu and this is marked by obvious 
Babylonianisms.'^ 

With the Nimrud clay tablet is easily confused the Nimrud 
slab.^ This dates from 743 and is thus the earliest inscription 
from the reign. But its account is so brief that it is of but 
trifling value. It assists a little in conjecturing what is lost from 
the tablet and mention of an event here is naturally of value as 

iCf. Olmstead. Jour. Anier. Or. Soc, XXXIV. 357. 

21. 5, 9ff., 16, 22, 47. 

3DT. 3. Schrader, Abh. Berl. Akad. 1880, 15ff., with photograph. 
For the Babylonian character, cf. Rost, 11. 

4Layard, NR. II. 33. L. 17f. Schrader. KB. II. 2ff.; Rost, 42fl.; 
Oppert, Exped., 336; Smith, Disc, 271; Meissner, Chrestomathie, lOf.; 
Menant, 138ff. 



ASSYRIAN HISTORIOGRAPHY 35 

establishing a minimum date. But where both have preserved 
the same account, the tablet is the fuller, and, in general, better, 
even though it is so much later.^ 

lOther inscriptions, III R. 10, 3, the place list; 83-1-18, 215, Winck- 
ler, AOF. II. 3f.; painted fragments, Layard, Nineveh and Babylon, 140f. 



CHAPTER V 

SARGON AND THE MODERN HISTORICAL CRITICISM 

The scources for the reign of Sargon (722.705)^ have already 
been discussed in detail elsewhere. All that is here needed is a 
summary of results.^ They fall into three well marked groups. 
The first includes the earlv inscriptions of the reign, which are 
miscellaneous in character.^ The circumstances under which 
Sargon came to the throne are indicated by a tablet from the 
second year which is of all the more value in that it is not a 
formal annals or display inscription.^ The Nimrud inscription 
comes from Kalhu, the earliest capital of Sargon. Unfortunately, 
it is very brief and is not arranged in chronological order. Aside 
from the rather full account of Pisiris of Carchemish, sufficient 
to date the inscription soon after its capture, we have only the 
briefest of references, and its value would be nothing, could we 
only secure the original, perhaps the earliest edition of the An- 
nals, on which it is based.^ A brief fragment may be noted be- 
cause of its mention of the sixth year, though we cannot be sure 
of the class to which it belongs. *5 Other fragments are either un- 
published or of no importance."^ 



iCollected in Winckler, Keilschrifttexte Sargons, 1889. 

201mstead, Western Asia in the Days of Sargon of Assyria, 1908, 



Iff. 



sSargon, 17ff. 

4K. 1349; Winckler. Sammlung. II. 1; AOF. I. 401ff. 

•■"'L. 33f.; "Winckler, Sargon. I. 168ff.; II. 48; Lyon, Assyr. Manual, 
9f.; Peiser, KB. II. 34ff.; Menant, 204ff. 

6K. 1660; "Winckler, Sammlung, II. 4. 

7K. 221+2669; K. 3149; K. 3150; K. 4455; K. 4463, "Winckler, 
Sammlung, II. 6; K. 4471, ihid. II. 4; DT. 310; 83-1-18. 215. The un- 
published fragments known from Bezold, Catalogiie, ad loc. 

(36) 



ASSYRIAN HISTORIOGRAPHY 37 

As a proved source for the second group, the newly discov- 
ered tablet should begin our study. ^ From the standpoint of 
source study, it is of exceptional value as it is strictly contempo- 
raneous and yet gives a very detailed account in Annals form of 
the events of a single year. The tablet was "written", probably 
composed, though it may mean copied, by Nabu shallimshunu, 
the great scribe of the King, the very learned, the man of Sar- 
gon, the eldest son of Harmaki, — seemingly an Egyptian name, — 
and inhabitant of the city of Ashur. It was brought (before 
the God Ashur?) in the limmu or eponym year of Ishtar duri, 
714-713, and tells us of the events of 714. It is written on an 
unusually large tablet of clay and is in the form of a letter. It 
begins "To Ashur the father of the gods . . . greatly, greatly may 
there be peace. To the gods of destiny and the goddesses who 
inhabit E har sag gal kurkurra, their great temple, greatly, greatly 
may there be peace. To the gods of destiny and the goddesses 
who inhabit the city of Ashur their great temple, greatly, greatly 
may there be peace. To the city and its inhabitants may there 
be peace. To the palace which is situated in the midst may there 
be peace. As for^ Sargon the holy priest, the servant, who fears 
thy great godhead, and for his camp, greatly, greatly there is 
peace." So this looks like a letter from the king to the god 
Ashur, to the city named from him, and to its inhabitants. Yet 
it is a very unusual rescript, very different from those which have 
come down to us in the official archives, especially in the use of 
the third person in speaking of the king, while in the regular let- 
ters the first is always found. Further, in the body of the sup- 
posed letter, the king, as is usual in the official annals, speaks in 
the first person. 

However it may be with the real character of the "letter," 
there can be no doubt as to its great value. To be sure, we may 
see in its boast that in the campaign but six soldiers were lost a 
more or less severe stretching of the truth, but, at least in com- 

iThureau-Dangin, Relation de la Huiticme Campagne de Sargon, 
1912. 

2So Thureau-Dangin, ad loc. 



38 UNIVERSITV OP MISSOURI STUDIES 

parison with the later records, it is not only much fuller, but far 
more accurate. Indeed, comparison with the later Annals shows 
that document to be even worse than we had dared suspect. 

Comparison of the newly discovered inscription with the 
parallel passages of the broken prism B shows that this is simply 
a condensed form of its original. The booty seems to have been 
closely copied, but the topographical details are much abbreviated. 
The discovery of this tablet, while supplying the lacunae in 
Prism B, has made this part useless. But all the more clearly is 
brought out the superiority, in this very section, of the Prism 
over the later Annals. Naturally, we assume the same to be true 
in the other portions preserved, in fact, the discovery of the tab- 
let has been a brilliant confirmation of the proof long ago given 
that this was superior to the Annals. ^ Unfortunately but a part 
of these fragments has been published^ and the difficulties in the 
way of copying these fragments have made many mistakes.^ 
But a few of these fragments have as yet been translated or even 
discussed.* For all parts of the reign which they cover, save 
where we have the tablet, they are now clearly seen to be our 
best authorities, nearer in date to the events they chronicle and 
much freer from suspicion than the Annals. The most urgent 
need for the history of the reign is that the fragments which are 
still unpublished^ should be published at once with a collation of 
those previously given. Even a translation and examination of 
the fragments already published would mark a considerable ad- 
vance in our knowledge of the period." 

lOlmstead, Sargon. llff.. with reconstruction of the order of the 
various fragments, as against Prasek, OLZ. XII. 117, who sharply at- 
tacked me "iiber den historischen wert den Stab zu brechen." 

2Winckler, Sargori, II. 45ff.; cf. I. xif. Photograph, Ball, Light 
from the East, 185. Thureau-Dangin, op. cit., 76ff. 

3To judge by a comparison of Winckler's text with that prepared by 
King for Thureau-Dangin, I. c. 

'Winckler, Sargon. I. 186f.; AOF. II. 71ff.; Mitth. Torderas. Gesell., 
1898, 1, 53; Thureau-Dangin, I. c. 

5Cf. Bezold. ZA. 1889, 411 n. 1. 

«For detailed study of Prism B, cf. Olmstead, I. c. 



ASSYRIAN HISTORIOGRAPHY 39 

Very similar to Prism B is our other broken prism, A.^ 
Both were found at Nineveh^ and this of itself proves a date 
some distance from the end of the reign when Sargon was es- 
tablished at Dur Sharruken.3 Prism A is of much the same type 
as the other, in fact, when we see how the Ashdod expedition, 
begun in the one, can be continued in the other ,^ we are led to be- 
lieve that the two had a similar text. If, however, the Dalta 
episode in each refers to the same event, then they had quite dif- 
ferent texts in this part of the history. Which of the two is the 
earlier and more trustworthy, if they did not have identical texts, 
and what are their relative relations cannot be decided in their 
fragmentary state, but that they are superior to the Annals is 
clear. Like Prism B, Prism A is worthy of better treatment and 
greater attention than it has yet been given. 

The third group consists of the documents from about the 
year 707, which have come down to us inscribed on the walls of 
Sargon's capital, Dur Sharruken.^ The earliest document of this 
group is naturally the inscription of the cylinders which were de- 
posited as corner stones,^ indeed, it closely agrees with the deed 
of gift which dated to 714J The same inscription is also found 
on slabs. ^ It is the fullest and best account of the building of 
Dur Sharruken, and from it the other documents of the group 
seem to have derived their building recital. Nor are other phases 
of the culture life neglected, as witness, for example, the well 

iWinckler, /Sargron, II. 44; I. 186ff.; Untersuch. Altor. Gesch., 118t£.; 
Textbuch3, 41f.; Rogers, 329f.; G. Smith, Disc, 288fE. Boscawen, Bab. 
Or. Rec. IV. 118ff. The Dalta episode and the beginning and end are 
still untranslated. 

2G. Smith, Disc, 147. 

3Cf. Olmstead, Sargon, 14 n. 

4As in Winckler, Sargon, I. 186fl. 

5For discussion of this group, cf. Olmstead, Sargon, 6ff. 

6Place, Nineve, II. 291ff.; Oppert, Dour Sarkayan, llff.; I R. 36; 
Lyon, Keilschrifttexte Sargons, Iff. Winckler, Sargon, II. 43; Menant, 
199ff.; Peiser, KB. II. 38ff. Barta, in Harper, 59ff. 

7Cf. Olmstead, Sargon, 178f. 

sMenant, RT. XIII. 194. 



40 UNIVERSITY OF MISSOURI STUDIES 

known attempt to fix prices and lower the high cost of Hving by 
royal edict. 

The remaining inscriptions of the group are all closely re- 
lated and all seem derived from the Annals. The display in- 
scription gives the data of the Annals in briefer form and in 
geographical order. Numbers are very much increased, and its 
only value is in filling the too numerous lacunae of its original.^ 
Imperfect recognition of its character has led many astray.^ 
Other inscriptions of the group are incised on bulls, on founda- 
slabs, on bricks, pottery, and glass, or as labels on the sculptures. 
Save for the last, they are of absolutely no value for the historian 
as they simply abstract from the Annals. As for the Cyprus 
stole, its location alone gives it a factitious importance.-^ 

The one important document of the group, then, is the An- 
nals. That, with all its value, it is a very much over estimated 
document, has already been shown.* There are four recensions, 
some of which differ widely among themselves and from other 
inscriptions. For example, there are three accounts of the fate 
of Merodach Baladan. In one. he is captured;^ in the second he 
begs for peace ;6 in the third, he runs away and escapes.'^ Nat- 
urally, we are inclined to accept the last, which is actually con- 
firmed by the later course of events. 

But it is only when we compare the Annals with earlier docu- 
ments that we realize how low it ranks, even among official in- 

iBotta, Mon..de Nineve, 95ff.; Winckler, Sargon. II. 30ff.; I. 97ff. 
Onpert-Menant, Pastes de fiaroon.-JA. 1863fE.; Menant. ISOff.; Op- 
pert, RPi, IX. Iff.; Peiser, KB. II. 52ff. 

2The error in connecting Piru and Hanunu, for example, already 
pointed out by Olmstead, Sargon, 10, is still held by S. A. Cook, art. 
Philistines, in the new Encyclopedia Britannioa. 

3For full bibliography of the minor inscriptions, cf. Olmstead, Sar- 
gon, 6f. For others since found at Ashur, cf. KTA. 37-42; 71; MDOG. 
20, 24; 22, 37; 25, 28, 31', 35; 26, 22; 31, 47; Andra, Tempcl, 91ff.; Taf. 
XXI; Genouillac-Thureau-Dangin, RA. X. 83ff. 

*01mstead, Sargon, 3ff. 

^Display 133. 

6 Annals V. 

7Annals 349. 



ASSYRIAN HISTORIOGRAPHY 41 

scriptions. Already we have learned the dubious character of its 
chronology. The Assyrian Chronicle has "in the land" for 712, 
that is, there was no campaign in that year. Yet for that very 
year, the Annals has an expedition against Asia Minor! It is 
prism B which solves the puzzle. In the earliest years, it seems 
to have had the same chronology as the Annals. Later, it drops 
a year behind and, at the point where it ends, it has given the 
Ashdod expedition as two years earlier than the Annals.^ Even 
with the old data, it was clear that the Prism was earlier and 
therefore probably more trustworthy ; and it was easy to explain 
the puzzle by assuming that years "in the land" had been later 
padded out by the Annals, just as we have seen was done for Dan 
Ashur under Shalmaneser III. Now the discovery of the tablet 
of the year 714 has completely vindicated the character of Prism 
B while it has even more completely condemned the Annals as a 
particularly untrustworthy example of annalistic writing. 

In the first place, it shows us how much we have lost. The 
tablet has 430 lines, of which a remarkably small portion con- 
sists of passages which are mere glorifications or otherwise of 
no value. Out of this mass of material, the Annals has utilized 
but 36 lines. That this is a fair sample of what we have lost in 
other years is hardly too much to suspect. Further, it would 
seem that the Annals used, not the tablet itself, but, since it has 
a phrase common to the Annals and the Prism,^ but not found in 
the tablet, either the Prism itself or a common ancestor. 

The cases where we can prove that the editor of the Annals 
"improved" his original are few but striking. It is indeed 
curious that he has in a few cases lowered the numbers of his 
original, even to the extent of giving three fortified cities and 
twenty four villages^ where the tablet has twelve fortified cities 
and eighty four villages.'* On the other hand, by a trick espe- 
cially common among the Sargonide scribes, the 1,235 sheep of 

iCf. Olmstead, Sargon, 11. 

2Ann. 1251; Prism B, Thureau-Dangin, op. cit., 76f. 

3Ann. 105. 

*Tabl. 89. 



42 UNIVERSITY OF MISSOURI STUDIES 

the tablet^ has reached the enormous total of 100,225!- More 
serious, because less likely to be allowed for, is the statement that 
Parda was captured'' when the original merely says that it was 
abandoned by its chief.'* But the most glaring innovation of the 
scribe is where, in speaking of the fate of Rusash, the Haldian 
king, after his defeat, he adds "with his own iron dagger, like a 
pig. his heart he pierced, and his life he ended. "^ This has long 
been doubted on general principles,*' but now we have the proof 
that it is only history as the scribe would like it to have been writ- 
ten. For the new inscription, while giving the conventional 
picture of the despair of the defeated king, says not a word of 
any suicide."^ However, the tablet does elsewhere mention the 
sickness of Rusash,^ and it may well be that it is to this sickness 
that we must attribute his death later.^ The complete misunder- 
standing of the whole campaign by earlier writers^ ^ furnishes 
the clearest indication of the unsatisfactory character of our re- 
cital so long as we must rely entirely on the Annals. It is the 
discovery of conditions like these which forces us to subject our 
official inscriptions to the most rigid scrutiny before we dare use 
them in our liistory.^^ 

iTabl. 349. 

2Ann. 129; cf. Thureau-Dangin, op. cit., 68, n. 4 for comparison 
of numbers. The same phenomenon can be constantly seen in the 
huge increases of the numbers of the Display inscription as compared 
with its original, the Annals. 

3Ann. 106. 

4Tabl. 84. 

5Ann. 139. 

6Cf. Olmstead, Sargon, 111. 

7Tabl. 411ff. 

8/bid. 115. 

»Cf. Thureau-Dangin, op. cit., xix. 

loCompare, for example, the brief and inaccurate account in Olm- 
stead, Sargon, 112ff., with that in Thureau-Dangin, op. cit. on the basis 
of the new tablet. 

iiBotta, Monuments de Ninive, pi. 70ff.; 104fE.; 158ff.; Winckler, 
Sargon II. pi. Iff. Oppert in Place, Ninive, II. 309£E.; Les Inscriptions 
de Dour Sarkayan, 29£f.; RPi VII. 21fE.; Menant, 158ff.; Winckler, De 
inscriptione quae vocatur Annalium, 1886; Sargon, I. 3£f. 



CHAPTER VI 

ANNALS AND DISPLAY INSCRIPTIONS 

(Sennacherib and Esarhaddon) 

Of the sources for the reign of Sennacherib (705-686)/ 
the chief is the Annals, added to at intervals of a few years, and 
so existing in several editions. As usual, the latest of these, the 
Taylor inscription, has been accorded the place of honor, so that 
the earliest edition, the so called Bellino Cylinder, can be called 
by a well known historian "a sort of duplicate of" the Taylor 
inscription.2 As we have seen repeatedly, the exact reverse 
should be our procedure, though here, as in the case of Ashur 
nasir apal, the evil results in the writing of history are less 
serious than in the case of most reigns. This is due to the un- 
usual circumstances that, with comparatively few exceptions, 
there was little omission or addition of the earlier data. Reg- 
ularly, the new edition simply added to the old, and, as a result, 
the form of the mass of clay on which these Annals were writ- 
ten changes with the increased length of the document, the earlier 
being true cyHnders, while the latter are prisms.^ At the same 
time that the narrative of military events was lengthened, the ac- 
count of the building operations followed suit. A serious de- 
fect is the fact that these documents are dated, not by years, 
but by campaigns, with the result that there are serious ques- 
tions in chronology. The increase in the number of our editions, 
however, has solved many of these, as the date of the campaign 

iThe only fairly complete collection of sources for the reign is 
still Smith-Sayce, History of Sennacherih, 1878, though nearly all the 
data needed for a study of the Annals are given by Bezold, KB. II. 80ff. 
Extracts, Rogers, 340ff. Cf. also Olmstead, Western Asia in the reign 
of Sennacherib, Proceedings of Amer. Historical Assn., 1909, 94ff. 

2Maspero, Histoire, III. 273 n. 1. 

sKing, Cuneiform Texts, XXVI. 7f. 

(43) 



44 UNIVERSITY OF MISSOURI STUDIES 

can now usually be fixed by observing in which dated document 
it last occurs. 

Of the more than twenty five more or less complete docu- 
ments, the first is the so called Bellino Cylinder which dates from 
October, 702. The fact that it has been studied separately has 
tended to prevent the realization that it is actually only a recen- 
sion. As a first edition, it is a trifle fuller, but surprisingly 
little.^ Next comes Cylinder B, now represented by six complete 
and seven fragmentary cylinders. It includes campaign three and 
is dated in May, 700.2 Cylinder C dates from 697 and contains 
the fourth expedition.^ The mutilated date of Cylinder D may 
be either 697 or 695, but as it has one campaign more than Cyl- 
inder C of 697, we should probably date it to the latter year.* 
From this recension seems to have been derived the display in- 
scription recently discovered on Mt. Nipur, which was inscribed 
at the end of campaign five.^ 

Somewhat dififerent from these is the newest Sennacherib 
inscription,^ which marks the transition from the shorter to the 
longer cylinders."^ After the narrative of the fifth campaign, 
two others are given, and dated, not by the number of campaign 
as in the documents of the regular series, but by the eponyms, so 
that here we have actual chronolog>\ The two campaigns took 
place in 698 and 695 respectively, the inscription itself being 
dated in 694. That they are not dated by the campaigns of the 

iK. 1680. Grotefend, Abh. Gottingen. Gesell. 1850. L. 63f. Smith- 
Sayce, If., 24ff., cf. 43fE. Oppert, Exped. I. 297fE.; Menant, 225ff.; Tal- 
bot, JRAS. XVIII. 76ff.; Trans. Roy. Soc. Lit.. VIII, 369ff.; RPi, I. 
23ff. It is the Bl. of Bezold. 

2Smith-Sayce, 30, 70f., cf. 24, 43, 53; Evetts, ZA. III. ailff.; for 
list of tablets, cf. Bezold, I. c. 

3K 1674; Smith-Sayce, 14, 76, cf. 30, 43, 53, 73, 78. The A 2 of 
Bezold. 

4BM. 22,508; K. 1675; Smith-Sayce, 24, 30, 43, 5'3, 73, 79; King, 
Cuneiform Texts. XXVI. 38, cf. p. 10, n. 2. The A 8 of Bezold. 

sinscription at Hasanah (Hassan Agha?) King, PSBA. XXXV. 66ff. 

6BM. 103,000; King, Cuneiform Texts, XXVI; cf. Pinches, JRAS. 
1910. 387fr. 

TKing, op. cit., 9. 



ASSYRIAN HISTORIOGRAPHY 45 

king and that they are not given in the later editions is perhaps 
due to the fact that the king did not conduct them in person. ^ 
The occasion for this new edition is not to be found, however, 
in these petty frontier wars, but in the completion of the new 
palace, in the increase in the size of the city of Nineveh, in the 
building of a park, and in the installation of a water supply, as 
these take up nearly a half of the inscription. The recovery of 
this document has also enabled us to place in the same group two 
other fragments, now recognized as duplicates.^ 

At about the same time must be placed the various inscrip- 
tions on the bulls which were intended to decorate this new pal- 
ace. One contains only five expeditions,^ the other has a brief 
sketch of the sixth,"* but both have references to the enthrone- 
ment of the crown prince Ashur nadin shum in Babylon.^ Still 
another gives a very full account of the sixth expedition, but 
there is no mention of Ashur nadin shum.^ This dates very 
closely the inscriptions of the period. The new inscription was 
written in August of 694. At this time as well as when the in- 
scription was placed on Bull II, the news of the sixth expedition, 
that across the Persian Gulf to Nagitu, had not yet come in. 
When this arrived, a brief account was hastily compiled and 
added to Bull III. But before a fuller narrative could be pre- 
pared, news came of the capture of Ashur nadin shum, which took 
place, as we know, soon after the Nagitu expedition, seemingly 
in the beginning of November."^ The inscription on Bull IV 

iKing, op. cit., p. 10. 

2BM. 102, 996, King, Cuneiform Texts, XXVI. 38; cf. p. 15, n. 1; K. 
4492, ibid. 39, not a reference to Tarbisi, as Meissner-Rost, Bauinschrif- 
ten, 94f.; as is shown by King, p. 18 n. 1. 

sBull 2, Smith-Sayce, 3, 24, 30f., 43, 51f., 53, 67f., 73, 78f., 86. L. 
60ff. (Bull 1 occurs only Smith-Sayce, 3.) 

iBuU 3, Smith-Sayce, I. c, and also 88f. 

sSmith-Sayce, 30f. 

sBull 4, Smith-Sayce, 3f., 24, 32ff., 43, 51, 53, 65ff.; 73, 77ff., 89ff.; A. 
Paterson, Palace of Sinacherib, 5f.; Ill R. 12f.; L. 38ff. 

7Bab. Chron. II. 36ff.; for kat TasJiriti in line 40, cf. Delitzsch, 
Ghronik, ad loc. 



46 UNIVERSITY OF MISSOURI STUDIES 

accordingly had an elaborate narrative of the Xagitu expedition, 
but all mention of the captured prince was cut out. 

The last in the series of Annals editions is the Taylor Prism 
of 690, generally taken as the standard inscription of the reign, 
and substantially the same text is found on seven other prisms.^ 
As has already been made evident, this is of no value for the 
earlier parts of the reign, since for that we have much better data, 
but it ranks well up in its class as comparatively little has been 
omitted or changed. Slightly earlier than the Taylor Cylinder 
is the Memorial or Nebi Yunus inscription, now at Constan- 
tinople, wdiich ends about where the other does. Here and 
there, it has the same language as the Annals group, but these 
coincidences are so rare that we must assume that they are due 
only to the use of well known formulae. In general, it is an 
abridgement of earlier records, though a few new facts are 
found. But for the second half of the sixth expedition, the re- 
volt of Babylon, it is our best source. Xot only is it fuller than 
the Tavlor prism, it gives a quite different account in which it is 
not the king but his generals who are the victors. Yet curiously 
enough, in the seventh expedition the Taylor cylinder is fuller 
and better.2 

Here too we may discuss the Bavian inscription, the display 
inscriptions cut in the rock where began the irrigation works con- 
structed to carry water to the capital. In their historical por- 
tions, they parallel the last campaign of the Taylor Prism, though 

IBM. 91,032, often given in photograph, especially in the "Bible 
Helps.'' A good photograph, Rogers, 543; Hist. op. 353. I R. 37ff; 
Smith-Sayce, passi7n: Delitzsch, Lesestiicke*, 54ff. ; Abel-Winckler, 17fE. 
Hornung, Das Sechsseitige Prisma des Sanherib. 1878; Bezold, KB. II. 
SOff., with numbers of the duplicates; Oppert, Les his. Assyr. des Sar- 
gonides, 41£f.; Menant, 214ff.; Talbot. RPi, I. 33ff.; Rogers, RP2, VI. 
80£f.; Harper, 68ff. Here also seem to belong the fragments 79-7-8, 305; 
K. 1665; 1651; S. 1026, as their text inclines toward that of the Taylor 
Prism. 

21 R. 43; A. Paterson, Palace of Sinacherib. 3; Smith-Sayce, 7f., 
39f., 68f., 86f., 102ff., lllff., 127ff.; Bezold, KB. II. 118f.; cf. King, 
Cuneiform Texts. XXVI. p. 10 n. 1. Seen at Constantinople in 1907-1908. 



ASSYRIAN HISTORIOGRAPHY 47 

in such different fashion that they may be considered separate 
sources. They then add the final capture and destruction of 
Babylon, of which they are the only Assyrian authority.^ Here 
too may be mentioned the two fraginents from the later part of 
the reign, on which is based a later expedition of Sennacherib 
against Palestine,^ as well as a tablet which seems to be a draft 
of an inscription to be set up in Kirbit in commemoration of the 
flight of Merodach Baladan.^ 

To complete our study of the sources for the reign, the more 
specifically building inscriptions may be noted.^ The greater part 
of what we know concerning the building operations of the reign 
comes from the documents already discussed. Of the specifically 
building inscriptions, perhaps the most important is the New 
Year's House inscription from Ashur,^ and the excavations there 
have also given a good number of display inscriptions on slabs^ 
and on bricks,'^ as well as some building prisms.^ 

Esarhaddon (686-668),^ like the others of his dynasty, pre- 
pared elaborate Annals.^*^ It is a poetic justice rarely found in 
history that the man who so ruthlessly destroyed the Annals of 

iIII R. 14; Pognon, LHnscription de Bavian. 1879; Smith-Sayce, 
129ff.; 157; King, Tukulti Ninib. 114ff. Menant, Nineve et VAssyrie, 
234fl.; Pinches, RPi, IX. 21ff.; Bezold, KB. II. 116ff. The order of 
date is B, C, A, D, Meissner-iRost, Bauinschriften, 67. Squeezes were 
secured by the Cornell Expedition. 

2Smith-Sayce, 137f.; the later fragment, Scheil, OLZ. VII. 69f.; 
Ungnad, Vorderas. Denkmdler, I. 73ff. ; in Gressmann, I. 121; Rogers, 
3451 

3III R. 4, 4; Strong, JRAS. XXIII. 148ff. 

4Meissner-Rost, Bauinschriften StinJierihs, 1893. 

5MD0G. 33, 14. 

6KTA. 43ff., 73f.; MDOG. 21, 13ff.; 22, 17fE.; 26, 27fe.; 43, 31; 44, 29. 

71. R. 7, VIII. H; Bezold, KB. 1141; KTA. 46-49; 72; MDOG. 20, 
24; 21, 12ff.; 22, 15; 25, 36f. 

8MD0G. 21, 37; 25, 22f.; 47, 39. 

sinscriptions of the reign collected by Budge, History of Esarhad- 
don, 1880. 

loFirst reference, G. Smith, TSBA. III. 457. Boscawen, ibid. IV. 
84ff.; IIIR. 35, 4; Budge, 114fE. ; Rogers, Haverford Studies, 11. Winck- 
ler, Untersuch z. altor. Gesch., 971; Winckler, Textbuch, 52ff.; Ung- 



48 UNIVERSITY OF MISSOURI STUDIES 

Tiglath Pileser IV is today known to us by still smaller frag- 
ments of his own. Aside from five mutilated lines from the 
ninth expedition, only a part of the first expedition against Egypt 
has survived and that in a very incomplete manner. We are 
accordingly dependent for our knowledge of the reign on the 
display inscriptions, with all their possibilities for error, and only 
the Babylonian Chronicle gives a little help toward fixing the 
relative order of events. 

The greater part of the history of the reign must be secured 
from the three most important cylinders. A and C are complete 
and are practically identical.^ B is broken and was originally 
considerably fuller, but seems to be from the same general series.^ 
The date of all three is probably 673.^ In comparing the texts 
of A-C and B, we note that in the first part, there seem to be no 
important differences, save that B adds an account of the acces- 
sion. In the broken part before this, B must have given the in- 
troduction and the murder of Sennacherib. Computation of the 
minimum in each column of B, based on the amount actually 
preserved in A and C, will give us some idea of what has been 
lost. Column II of B must have been devoted in part to the 
final defeat of the rebels and in part to the introduction to the 

nad, I. 123; Rogers, 357ff. Cf. also G. Smith, Disc. SllfE.; Delattre. 
L'Asie, 149; Olmstead, Bull. Amer. Oeog. Soc. XLIV. 1912, 434. 

148-10-31, 2; L. 20ff.; I R. 45ff.; Abel-Winckler, 22ff.; Budge. 32ff.; 
Harper, Weftraico. III. 177ff.; IV. 99fE. Abel, KB. II. 124ff.; Oppert, Ins. 
des Sargonides, 53ff.; Talbot, Jour. Sacr. Lit., IX. 68ff.; Trans. Roy. 
Soc. Lit., VII. 551ff.; RPi, III 109ff.; Menant, 241fE.; Harper, 81fE. 
C was used by R. for restoring A. Text, Harper, Hcbraica, IV. 18ff., 
with the parallels 80-7-19, 15, and K. 1679. Also King, Supplement, 108f. 

248-11-4, 315; III R. 15f.; Budge, 20ff.; 97fE.; Harper, Hebraica, 
III. 177£f.; IV. 146ff.; Abel-Wincl<ler, 25f. Winckler, KB, II. 140ff. Har- 
per, 80f.; Menant, 248ff.; Talbot, RPi. III. 102ff.; North Brit. Rev., ISIO, 
quoted Harper, Hebr. I. c. 

3C is dated in the month Abu, cf. Harper, Hebr. IV. 24; B, accord- 
ing to Budge, ad loc, has Abu of the year 673, but Winckler, I. c. omits 
the month. If the month is to be retained, the identity of month points 
to identity of year, and there is nothing in B to prevent this con- 
jecture. A is from Nebi Yunus, B from Koyunjik. 



ASSYRIAN HISTORIOGRAPHY 49 

long narrative concerning Nabu zer lishir. As at least four lines 
were devoted to this introduction in the usually much shorter D, 
it must have been fairly long in B. Why A omitted all this is a 
question. That these two events are the first in the reign is made 
clear by the Babylonian Chronicle, so that thus far the chron- 
ological order has been followed. The next event in B and the 
first in A is the story of the Sidon troubles, and again the Chron- 
icle shows it to be in chronological order. Since A has no less 
than 49 lines to deal with the events in the lost beginning of 
column III, it is clear that the much fuller B has here lost much. 
In the gap in Column IV, we are to place the Aduma narrative 
and the traces where we can begin to read show that they are in 
the conclusion of the Median troubles. ^ For the lost part of the 
fifth column, we must count the ladi and Gambulu expeditions, 
and a part of the building narrative. About the same building 
account as in A must be placed at the commencement of column 
VI. The irregularity in the minimum numbers for the difiFerent 
columns, on the basis of A, shows that B had in some cases much 
longer accounts than in others, and this is confirmed where B 
gives a complete list of Arabian and of Syrian kings while A 
does not. These minimum numbers also indicate that but about 
one-fourth of B has been preserved. However, the over lapping 
gives us some reason to hope that nearly all its facts have been 
preserved in the one or the other edition. 

We have already seen that strict chronology is followed by 
B, strange to relate, in the order, punishment of the assassins, 
681, Babylon, 680, and Sidon, 677. Then A gives the Kundu 
troubles which, according to the Chronicle, follow in 676, and 
Arzani and the brook of Egypt, which fit well enough with the 
Egyptian expedition given under 675. These are the only sec- 
tions we can date chronologically, and the order is chrono- 
logically correct. But whether we can assume this for all the 
events mentioned may be doubted in the light of the disagreement 

^Shevashun of B. is the shepushshim of A. IV. 36, and the elishun 
ukin is virtually the same as ukin sirushun. 
4 



50 UNIVERSITY OF MISSOURI STUDIES 

between A and B in their order. In placing the Arabs before 
Bazu, or the Babylonian Nabu zer lishir before Bit Dakkuri, A 
is clearly attempting a more geographical order. We shall then 
use B as our main source whenever preserved, supplemented by 
A when the former is missing, but we must not forget that all 
are simply display inscriptions. 

Another display inscription of the same type we shall call 
D. It is close to B as is shown in the story of Nabu zer lishir, 
is seemingly briefer than that document, but is certainly fuller 
than A, and is independent of both. The order of events is Baby- 
lon, Egypt, Hubushna. As D omits Sidon and the Cilician cities, 
found in one of the others and proved to the period by the Baby- 
lonian Chronicle, it is clear that we have here only extracts, even 
though the events narrated are given more fully than in A.^ 
Still another document of similar character may be called E. As 
it mentions the Uabu rebellion which is not in A, it should date 
after 673, and its order, Chaldaeans, Gambulu, Egypt, Arabs, 
Sidon, Asia Minor, is not chronological but geographical. It has 
some striking variants in the proper names, for example, we have 
here Musur, universally recognized as meaning Egypt, where 
A has Musri, and thus we have exact proof that Musri does 
equal Egypt, the advocates of the Musri theory, if any still sur- 
vive, to the contrary notwithstanding. ^ It is also longer than A 
in the River of Eg}'pt section, and than B in the Elam account. 
As a late document, it is of value only for the Uabu affair.^ 
We may also note here another prism fragment* and a slab with 
a brief account of many campaigns. The first, that against Bazu. 
we know dates to 676. The others, to Uruk, to Buesh king of 
an unknown land. Akku, and the king of Elam, are of doubtful 
date, but are almost certainly later.' 

iK. 2671; Winckler, ZA. II. 299ff.; AOF. I. 522. 

2Cf. Olmstead, Sargon, 56fl. 

3Winckler, ZA. II pi. II; AOF. I. 526fE. 

480-7-19, 15; Winckler, Untersuch. z. altor. Gesch., 98. Cf. King, 
Supplement, 109. 

5K. 8544; Winckler, AOF. I. 532.— I have been unable to see Scheil, 
Le Prisme S d'Assarhaddon. 



ASSYRIAN HISTORIOGRAPHY 51 

Finally, we must discuss two display inscriptions from the 
very end of the reign, whose importance is in no small degree 
due to the locality in which they were found. One is the famous 
stele discovered amid the ruins of the North Syrian town of 
Sinjirli. It dates after the capture of Memphis, 671, and seems 
to have been composed on the spot, as it shows no relationship 
to other inscriptions. ^ The same is probably true of the equally 
famous rock cut inscription at the Dog River (Nahr el Kelb), 
north of Berut. Though the oldest Assyrian inscription to have 
a cast taken, it seems never to have been published. It is rapidly 
disappearing, as the fact that it was cut through a very thin 
layer of hard rock has caused much flaking. Esarhaddon is called 
King of Babylon and King of Musur and Kusi, Egypt and Ethi- 
opia, and the expedition against Tarqu, which ended with the 
capture and sack of Memphis, is given. Thus it agrees with the 
Sinjirli inscription and may well date from the same year.^ 

We have a considerable number of building inscriptions, but 
there are few source problems in connection with them.^ Per- 
haps the most important is the prism which tells so much in 
regard to the earliest days of Assyria.^ Another important 
document is the Black Stone, a four sided prism with archaistic 
writing. It was found at Nineveh, though it deals with the re- 
building of Babylon, and seems to date from the first year.^ Two 

iPhotograph and text, Schrader, in Luschan, Ausgrabungen in 
SendscMrli, I. llff., and pi., cf. Rogers, 551; Hist, op. 399; Paterson, 
Sculptures, 103. Harper, 90fE. I have been able to consult squeezes in 
the library of Cornell University. 

2Translation, G. Smith, Epom/m Canon. 167ff. The text, so far as I 
know, has never been published, even in connection with the elaborate 
study of the Nahr el Kielb sculptures by Boscawen, TSBA. VII. 345. I 
have been able to use the squeeze taken in 1904 in connection with 
Messrs. Charles and Wrench, but much less can now be seen than what 
Smith evidently found on the cast. Cast, Bonomi, Trans. Roy. Soc. Lit., 
III. 105; Nineveh and its Palaces, 5f. 86. 142fE., 367. 

3Collected in Meissner-Rost, Beitr. z. Assyr., III. 189ff; Thureau- 
Dangin, Rev. Assyr. XI. 96ff. 

4KTA. 51; MDOG. 25, 33. 

51 R. 49; Winckler, KB. II. 120ff.; Meissner-Rost, 218fC. Oppert, 



52 UNIVERSITY OF MISSOURI STUDIES 

Others date after 675 as the one on a stone slab from the south 
west palace at Kalhu states that he took captive the king of 
Meluh/ and the other stone tablet gives him Egyptian titles,^ so 
that they must be placed after the capture of that country. We 
may also mention in conclusion the one v^'hich gives the restora- 
tion of the Ishtar temple at Uruk"^ and the various ones found at 
Ashur by the German excavators.^ 

Exped.. I. 180 f.; Menant, 248; Babylone et Chaldre, 167f.; Harper, 88f. 
King, Supplement, 38, dates from Aru of accession year. 

iL. 19a. Winckler, KB. II. 150f. Oppert. Bxped.. I. 324; Menant, 240. 

21 R. 48, 5; Winckler, KB. II. 150f.; Meissner-Rost, 204ff.; Menant, 
249. 

381-6-7, 209; Winckler, KB, II. 120 n. 1; Barton, Proc. Amer. Or. 
Soc, 1891, cxxx. 

4KTA. 51-55; 75; MDCK5. 20, 26ff.; 22, 12f.; 25, 33, 65; 26. 20f.; 
26, 41ff.; 28. 13, 49, lOf. Weissbach, in Koldewey, Die Tempe/ von Bcbi/- 
lon, 71. 



CHAPTER Vn 

ASHUR BANI APAL AND ASSYRIAN EDITING 

The reign of Ashur bani apal (668-626), stands preeminent 
for the mass of material available, and this has twice been col- 
lected.'- Yet in spite of all this, the greater number of the in- 
scriptions for the reign are not before us in adequate form, and 
there are problems which only a renewed study of the originals 
can solve. 

Once again we have the usual Annals as our main source. 
Earlier scholars have in general satisfied themselves with the 
publication and study of the latest edition, sometimes supple- 
mented by more or less full extracts from the others. There are 
reigns, such as that of Sennacherib, where such procedure re- 
sults in comparatively little distortion of the history. But in no 
reign is the distortion of the earlier statements more serious, in- 
deed one can hardly recognize the earlier documents in their later 
and "corrected" form. Accordingly, in no reign is it more im- 
perative that we should disentangle the various sources and give 
the proper value to each. When we have discovered which docu- 
ment is our earliest and most authentic source for any given 
event, we have already solved some of the most stubborn prob- 
lems in the history of the reign. The various conflicting accounts 
of the Egyptian campaigns, for example, have caused much 
trouble, but if we recognize that each is a step in the movement 
toward increasing the credit the king should receive for them, 
and trust for our history only the first in date, we have at last 
placed the history of the reign on a firm basis. 

Our very earliest document furnishes a beautiful illustra- 
tion of this principle. It is a detailed narrative of the unimpor- 

iG. Smith, History of Assur'banipal, 1871; S. A. Smith, Keilschrift- 
texte Asurbanipals, 1887ff. 

(53) 



54 UNIVERSITY OF MISSOURI STUDIES 

tant Kirbit expedition, which is ascribed to the governor Nur 
ekalli umu. Cyhnder E gives a briefer account and Cylinder F 
one still shorter. Both vaguely ascribe it to the "governors" but 
do not attempt to claim it for the king. It remained for Cylinder 
B, a score of years later, to take the final step, and to inform us 
that the king in person conducted the expedition. Further, the 
formal conclusion, which immediately follows the Kirbit expedi- 
tion in our earliest document, shows that this event, unimportant 
as it was, was the only one which could be claimed for the "be- 
ginning of the reign." This campaign is further fixed by the 
Babylonian Chronicle to the accession year. Yet later cylinders 
can place before it no less than two expeditions against Egypt and 
one against Tyre ! Our earliest document alone would be enough 
to prove that these had been taken over from the reign of his 
father, even did we not have some of this verified by that father 
himself.^ 

Next in date and therefore in value we are probably to place 
Cylinder E, a decagon fragment, which contains a somewhat less 
full account of the Kirbit campaign, and a picturesque narrative 
of the opening of diplomatic relations with Lydia. Before these 
events, it placed an account of the Egyptian expedition. Al- 
though only a portion is preserved, it is sufficient to show that the 
"first Egyptian expedition" at least was credited to his father. ^ 

A third account, which we may call F, gave credit for the 
earlier half of the Egyptian campaigns to his father and for the 
latter half to his own lieutenants. The references to Tabal and 
Arvad indicate that some time had elapsed in which memorable 
events in his own reign could have taken place, and this is con- 
firmed by the much more developed form of the Lydian narra- 
tive, with its dream from Ashur to Gyges. and its order for ser- 
vitude. That this account is of value as over against the later 
ones has been recognized,*^ but we should not forget that it al- 

iK. 2846; Winckler, AOF. I. 474ff. 

2G. Smith, 34f., 76f., 82f.; K. 3083 is identical for a line each with 
Cyl. E and P. 

3Tiele, Gesch. 372. 



ASSYRIAN HISTORIOGRAPHY 55 

ready represents a developed form of the tradition. ^ Somewhat 
later would seem to be the account we may call G. Here the 
Egyptian wars are still counted as one expedition, but a second 
has been stolen for Ashur bani apal by taking over that campaign 
of his father against Baal of Tyre which is given in the Sinjirli 
inscription.2 

With Cylinder B, we reach the first of what is practically 
a new series, so greatly has the older narrative been "corrected" 
in these later documents. Both the Egyptian wars have now 
been definitely assigned to the king, and the making of two ex- 
peditions into Egypt has pushed the one against Baal of Tyre up 
to the position of third. The octagon B dates from the midst of 
the revolt of Shamash shum ukin and is a most highly "corrected" 
document.^ 

The story of the Shamash shum ukin revolt is continued by 
Cylinder C, a decagon, whose form points to the fact that it is a 
fuller edition. In general, its text holds an intermediate position 
between A and B, the lists of Syrian and Cypriote kings, which 
are copied verbatim from the Cylinder B of Esarhaddon,^ being 
found only in it.^ With C should in all probability be listed two 
decagons one of which is called Cylinder D.*^ Then comes a 
document which we may call H, with several duplicates, and as 
the Ummanaldas episode is dealt with in fuller form than in A, 

iK. 2675; III R. 28f.; G. Smith, 36ff., 56ff., 73ff., 80ff.; cf. 319 and 
S. A. Smith, II. 12fE., for ending giving erection of moon temple at 
Harran, a proof that we have the conclusion and so can date approxi- 
mately; Winckler, UntersucJi. z. altor. Gesch., 102ff.; Jensen, KB. II. 
236ff. A fragmentary stone duplicate from Babylon, Delitzsch, MDOG., 
XVII 2 n.* 

2K. 3402; G. Smith, 78. 

3G. Smith, passim; Jensen, KB. II. 240ff.; Menant, 278ff.; for the 
duplicate K. 1729 from which most of the B text is taken, cf. Johns, 
PSBA. XXVII. 97. 

4V. 13ff. 

5Rm. 3; G. Smith, 30ff., 178ff., cf. 15, 52, 151, 319; S. A. Smith, II. 
25fE.; Menant, 277f. Jensen, KB. II. 238ff., 266ff. 

6G. Smith, 317f. K. 1794; III. R. 27a; S. A. Smith, II. 18, cf. G. 
Smith, 319. 



56 UNIVERSITY OF MISSOURI STUDIES 

it probably dates earlier.' For the Tamaritu events, we have a 
group of tablets of unknown connections.^ 

All the documents thus far considered are fuller and more 
accurate in dealing with the events they narrate than is the group 
which has so long been considered the standard. The first known 
was Cylinder A, a decagon, whose lines divide the document into 
thirteen parts. It is dated the first of Nisan (March) in the 
eponymy of Shamash dananni, probably 644.'^ Earlier scholars 
made this the basis of study, but it has since been supplanted by 
the so called Rassam cylinder, a slightly better preserved copy, 
found in the north palace of Nineveh, and dated in Aru (May) 
of the same year.^ Still a third is dated in Ululu (September) 
of this year.^ 

That this document is by no means impeccable has long been 
recognized. Already George Smith had written "The contempt 
of chronology in the Assyrian records is well shown by the fact 
that in Cylinder A, the account of the revolt of Psammitichus is 
given under the third expedition, while the general account of the 
rebellion of [Shamash shum ukin] is given under the sixth expe- 
dition, the aflfair of Nebobelzikri under the eighth expe- 
dition, and the Arabian and Syrian events in connection are given 
under the ninth expedition. "^^ If this severe criticism is not 
justified by a study of the Assyrian sources as a whole, the ref- 
erence to Cylinder A may well begin our consideration of the 
shortcomings of that group. The Karbit and Urtaki episodes 

iK. 2656; G. Smith, 215ff. Are the duplicates mentioned here to 
be found in K. 2833 and K. 3085. G. Smith. 205? 

2K. 1364; 3062; 2664; 3101; 2631'; G. Smith, 243ff.— Where we 
are to place the cylinder Rm. 281, dealing with Urtaki's reign, Winck- 
ler, AOF. I. 478 n. 2, cannot be told until it is published. 

3G. Smith, passim. III R. 17ff. RPi, IX 37fe.; Menant. 253ff. 

*BM. 91,026; Rm. 1; Photograph. Rogers, 555; ffis^ op. 444. V. "R. 
1-10; Abel-Winckler, 26ff.; Winckler, Sammlung, III; S. A. Smith, I. 
Jensen, KB. II. 152fE. J. M. P. Smith, in Harper, 94ff.; Lau & Lang- 
don, Annals of Ashurhanapal, 1903. 

5G. Smith. 316. 

6Ibid. 202 n.* 



ASSYRIAN HISTORIOGRAPHY 57 

are entirely omitted. The omission of Karbit has dropped the 
Manna from the fifth to fourth and the omission of the latter 
has made the Teumman campaign the fifth instead of the seventh 
as in B, while the Gambulu expedition is also listed in the fifth 
though B makes it the eighth! The death of Gyges is added im- 
mediately after the other Lydian narrative, without a hint that 
years had intervened. The elaborate account of Teumman given by 
B has been cut decidedly and the interesting Ishtar dream is en- 
tirely omitted. 

The same is true of the Gambulu narrative. While B and 
C have the data as to the Elamite side of the revolt of Shamash 
shum ukin, the introduction and conclusion as well as many new 
details are found only in A. It is curious to find here, for 
the first time, the greater part of the long list of conquered 
Egyptian kings, written down when Egypt was forever freed 
from Assyrian rule. That Cylinder B was not its immediate 
source is shown by the fact that in the first Egyptian expedition 
it gives the pardon of Necho, which is not in B, but is found in 
the earlier F. 

Although this document has regularly been presented as the 
base text, largely because it gives a view of the greater part of 
the reign, enough should have been said in the preceding para- 
graph to prove how unworthy of the honor it is. Of all the cases 
where such procedure has caused damage, this is the worst. For 
the years from which we have no other data, we must use it, and 
we may hope that, as this period was nearer the time of its ed- 
itors, its information may here be of more value. But we should 
recognize once and for all that the other portions are worthless 
and worse than worthless, save as they indicate the "corrections" 
to the actual history thought necessary by the royal scribes. 

Later than this in date, in all probabiHty, is the document we 
may call I. To be sure, the Arabian expedition already occurs 
in B, but I has also sections which appear only in A, and which 
therefore probably date later. The one indication that points to 
its being later than A is the fact that, while A ascribes these ac- 
tions to his generals, our document speaks of them in the first 



58 UNIVERSITY OF MISSOURI STUDIES 

person.^ Still later are the Beltis^ and Nabu inscriptions,' 
though as these are merely display inscriptions, the date matters 
little. Here too belongs J in spite of its references to the ac- 
cession.* And to this very late period, when the empire was fall- 
ing to pieces, is to be placed the hymn to Marduk which speaks 
of Tugdami the Cilician.^ 

We have already crossed the boundary which divides the 
really historical narratives from those which are merely sources. 
Among the latter, and of the more value as they open to us the 
sculptures, are the frequent notes inscribed over them,^ while a 
number of tablets give much new historical information from the 
similar notes which the scribe was to thus incise." The Ishtar 
prayer is a historic document of the first class, the more so as 
its author never dreamed that some day it might be used to prove 
that the king was not accustomed, as his annals declare, to go 
forth at the head of his armies, that he was, in fact, destitute of 
even common bravery.^ 

For the period after the reign of Ashur bani apal, we have 
only the scantiest data. The fall of the empire was imminent 
and there were no glories for the scribe to chronicle. Some bricks 
from the south east palace at Kalhu,^ some from Nippur,^*^ and 
some boundary inscriptions^^ are all that we have from Ashur 

iK. 2802; G. Smith, 290ff. 

211 R. 66; G. Smith 303ff.; S. A. Smith, II. lOff.; cf. I. 112; Jen- 
sen, KB. II. 264ff.; Menant, 291ff. 

3S. A. Smith, I. 112ff.; III. 128ff.; Strong, RA. II. 20ff. 

*K. 2867; S. A. Smith, II. Iff.; cf. Olmstead, BuU. Amer. Oeog. Soc, 
XLIV. 434. — The various British Mueeum fragments, cited in King, Sup- 
plement, seem to be of no special importance for this study as they 
are duplicates with few variants. 

sS. A. Strong, JA. 1893, 1. 368ff. 

^Scattered through the work of G. Smith, cf. also Menant, 287ff. 

7K. 2674: III R. 37; G. Smith, 140ff.; S. A. Smith, III. Iff. K. 4457; 
G. Smith, 191ff. K. 3096; G. Smith, 295ff. 

«K. 2652; III R. 16, 4; G. Smith, 139f.; S. A. Smith, III. llff.; cf. 
Jensen, KB. II. 246ff. Talbot. TSBA. I. 346ff. 

91 R. 8, 3; Winckler, KB. II. 268f.; Menant, 295. 

loHilprecht, ZA. IV. 164; Explorations. 310. 

iiK. 6223, 6332; Winckler, AOF. II. 4f; Johns. PSBA. XX. 234. 



ASSYRIAN HISTORIOGRAPHY 59 

itil ilani and from Sin shar ishkun only fragments of a cylinder 
dealing with building.^ We have no contemporaneous Assyrian 
sources for the fall of the kingdom, our only certain knowledge 
being derived from a mutilated letter^ and from a brief statement 
of the Babylonian king Nabu naid a generation later.'' 

IK. 1662 and dupl. I R. 8, 6; Schrader, SB. Berl. Gesell. 1880, Iff.; 
Winckler, Rev. Assyr. II. 66ff.; KB. II. 270ff.; MDOG. XXXVIII. 28. 
2BM. 51082; Thompson, Late Babylonian Letters 248. 
3Messerschmidt, Mitth. Vorderas. Gesell., 1896. I. 



CHAPTER Vm 

THE BABYLONIAN CHRONICLE AND BEROSSUS 

This concludes our detailed study of the "histories" of the 
reigns which were set forth with the official sanction. Before 
summing up our conclusions as to their general character, it will 
be well to devote a moment to the consideration of certain other 
sources for the Assyrian period. Many minor inscriptions have 
been passed by without notice, and a mere mention of the mass 
of business documents, letters, and appeals to the sun god will 
here be sufficient, though in a detailed history their help will be 
constantly invoked to fill in the sketch secured by the study of 
the official documents, and not infrequently to correct them. Of 
foreign sources, those of the Hebrews furnish too complicated a 
problem for study in this place,^ and the scanty documents of the 
other peoples who used the cuneiform characters hardly furnish 
source problems. 

Even the Babylonians have furnished us with hardly a text 
which demands source study. To the end, as is shown so con- 
spiciously in the case of Nebuchadnezzar, scores of long inscrip- 
tions could be devoted to the building activities of the ruler while 
a tiny fragment is all that is found of the Annals. Even his 
rock cut inscriptions in Syria, those in the Wadi Brissa and at 
the Nahr el Kelb, are almost exclusively devoted to architectural 
operations in far away Babylon \^ 

Yet if the Babylonians were so deficient in their apprecia- 
tion of the need of historical annals for the individual reigns, 
they seem to have been the superiors of the Assyrians when it 
came to the production of actual histories dealing with long 

iCf. Olmstead, AJSL. XXX. Iff.; XXXI. 169fE.. for introduction to 
these new problems. 

2lt may be noted that the Cornell Expedition secured squeezes of 
both these inscriptions. 

(60) 



ASSYRIAN HISTORIOGRAPHY 61 

periods of time. While the Babylonians have preserved to us 
numerous lists of kings and two excellent works which we have 
every reason to call actual histories, the Babylonian Chronicle 
and the Nabunaid-Cyrus Chronicle, the Assyrians have but the 
Eponym Lists, the so called Assyrian Chronicle, and the ,so 
called Synchronous History. The last has already been dis- 
cussed, and we have seen how little it deserved the title of a real 
history, yet it marks the greatest advance the Assyrians made 
along this line. The Eponym lists are merely lists of the officials 
who dated each year in rotation, and they seem to have been com- 
piled for practical calendar purposes. The so called Assyrian 
Chronicle is in reality nothing but a chronological table in three 
columns, the first with the name of the eponym for the year, the 
second with his office, and the third with the most important 
event, generally a campaign, of the year. As a historical source, 
more can be made out of this dry list than has previously been 
suspected, and this has been pointed out elsewhere.^ But, as a 
contribution to the writing of history, it holds a distinctly low 
place. 

On the other hand, the Babylonian Chronicle is a real, if 
somewhat crude history. In fact, it can be said without fear of 
contradiction that it is the best historical production of any 
cuneiform people. Our present copy is dated in the twenty sec- 
ond year of Darius I of Persia, 500 B. C, but, as it was copied 
and revised from an earlier exemplar, which could not always 
be read, its original must be a good bit earlier. Only the first 
tablet has come down to us, but the mention of the first proves 
that a second existed. What we have covers the period 745-668, 
a period of seventy-seven years. The second tablet would cover 
a period nearer the time of the writer and would naturally deal 
with the events more in detail, so that a smaller number of years 
would be given on this tablet. If but two tablets were written, 
the end of the work would be brought down close to the time when 
the Assyrian Empire fell (608). It is a tempting conjecture, 
though nothing more, that it was the fall of Assyria and the in- 

lOlmstead, Jour. Amer. Or. Soc, XXXIV. 344fl. 



62 UNIVERSITY OF MISSOURI STUDIES 

terest in the relations between the now dominant Babylonia and 
its former mistress, excited by this event, which led to the com- 
position of the work. Be that as it may, the author is remark- 
ably fair, with no apparent prejudice for or against any of the 
nations or persons named. The events chosen are naturally al- 
most exclusively of a military or political nature, but within 
these limits he seems to have chosen wisely. In general, he con- 
fines himself to those events which have an immediate bearing on 
Babylonian history, but at times, as, for example, in his narration 
of the Egyptian expeditions, he shows a rather surprising range 
of interest. If we miss the picturesque language which adds so 
much to the literary value of the Assyrian royal annals, this can 
hardly be counted an objection by a generation of historians 
which has so subordinated the art of historical writing to the 
scientific discovery of historical facts. In its sobriety of presen- 
tation and its coldly impartial statement of fact, it may almost 
be called modern.^ 

We know the name of our other Babylonian historian, and 
we also know his date, though unfortunately we do not know his 
work in its entirety. This was Berossus, the Babylonian priest, 
who prepared a Babyloniaca which was dedicated to Antiochus 
I. When we remember that it is this same Antiochus who is 
the only one of the Seleucidae to furnish us with an inscription 
in cuneiform and to the honor of one of the old gods,- it becomes 
clear that this work was prepared at the time when fusion of 
Greek and Babylonian seemed most possible, and with the de- 
sire to acquaint the Macedonian conquerors with the deeds of 
their predecessors in the rule of Babylonia. The book was char- 
acteristically Babylonian in that only the last of the three books 

iPhotograph, Rogers, 515. C. T. XXXIV 43ff. Abstract, Pincbes, 
PSBA. VI. 19Sff. Winckler, ZA. II. 148ff.; Pinches, JRAS. XIX. 655ff. 
Abel-Winckler, 47f. Duplicates, Bezold, PSBA. 1889, 181; Delitzsch, 
Lesestiicke*, lS7ff. Scbrader, KB. TI. 274ff.: Delitzsch. Bab. Chronik; 
Rogers, 208ff. ; Barta, in Harper, 200ff. Sarsowsky, Keilschriftliches Ur- 
kundenbuch, 49ff.; Mercer, Extra Biblical Soui-ces, 65ff. 

2Best in Weissbach, Achdmeniden Inschriften, 132ff., cf. xxx for 
bibliography. 



ASSYRIAN HISTORIOGRAPHY 63 

into which it was divided, that beginning with the time of 
Nabonassar, can be considered historical in the strictest sense, 
and even of this only the merest fragments, abstracts, or traces, 
have come down to us. And the most important of these frag- 
ments have come down through a tradition almost without 
parallel. Today we must consult a modern Latin translation of 
an Armenian translation of the lost Greek original of the Chron- 
icle of Eusebius,^ who borrowed in part from Alexander Poly- 
histor who borrowed from Berossus direct, in part from Aby- 
denus who apparently borrowed from Juba who borrowed from 
Alexander Polyhistor and so from Berossus. To make a worse 
confusion, Eusebius has in some cases not recognized the fact 
that Abydenus is only a feeble echo of Polyhistor, and has quoted 
the accounts of each side by side! And this is not the worst. 
Although his Polyhistor account is in general to be preferred, 
Eusebius seems to have used a poor manuscript of that author. 
Furthermore, there is at least one case, that of the name of one 
of Sennacharib's sons, which can be secured only by assuming a 
mistake in the Armenian alphabet. 

It is in Eusebius that we find our most useful information, 
some of the facts being very real additions to our knowledge. 
But Berossus was also used by the early ApoUodorus Chronicle, 
some time after 144 B. C, from which some of his information 
may have drifted into other chronological writings. Alexander 
Polyhistor was used by Josephus, and Abydenus by Cyrillus, Syn- 
cellus, and the Armenian historian, the pseudo Moses of Cho- 
rene. So in these too, or even in others not here named, may lurk 
stray trifles from the work of Berossus. Perhaps from this, or 
from a similar source, comes the Babylonian part of the list of 
Kings known as the Canon of Ptolemy, which begins, as does the 
Babylonian Chronicle, with the accession of Nabonassar.^ 
Though directly of Egyptian origin, as is shown by the system of 

lA. Schoene, Eusebii Chronicorum libri duo, 1866ff. ; cf. Rogers, 
Parallels, 347ff.; J. Karst, Eusebius Werke, V. 

zThe most convenient edition Wachsmuth, Einleitung in das Stud- 
ium der alien Geschichte, 304fE.; cf. Rogers, 239. 



64 UNIVERSITY OF MISSOURI STUDIES 

dating, it undoubtedly goes back to a first class Babylonian source, 
as do the astronomical data in the Almagest of the same 
author, though here too the Egyptian calendar is used.^ Summing 
up, practically all the authentic knowledge that the classical 
world has of the Assyrians and Babylonians came from Beros- 
SUS.2 Herodotus may furnish a bit and something may be se- 
cured from the fragments of the Assyriaca of Ctesias, but it is 
necessary to test each fact from other sources before it can be 
accepted. 

And now what shall we say by way of summing up the Assy- 
rian writing of history? First of all, it was developed from the 
building inscription and not from the boast of the soldier. That 
this throws a new light on the Assyrian character must be ad- 
mitted, though here is not the place to prove that the Assyrian 
was far more than a mere man of war. All through the develop- 
ment of the Assyrian historiography, the building operations play 
a large part, and they dominate some even of the so called An- 
nals. But once we have Annals, the other types of inscriptions 
may generally be disregarded. The Annals inscriptions, then, 
represent the height of Assyrian historical writing. From the 
literary point of view, they are often most striking with their bold 
similes, and that great care was devoted to their production can 
frequently be proved. But in their utilization, two principles 
must constantly be kept in mind. One is that the typical annals 
inscription went through a series of editions, that these later edi- 
tions not only omitted important facts but "corrected" the earlier 
recitals for the greater glory of the ruler, real or nominal, and 
that accordingly only the earliest edition in which an event is 
narrated should be at all used. Secondly, we should never for- 
get that these are official documents, and that if we can trust them 
in certain respects the more because they had better opportuni- 

iCf. Olmstead, Sargon. 34f. 

20f the literature on Berossus, we may quote here only Miiller, Frag- 
menta Historicorum Graecorum, II. 495ff. ; and the various articles by 
Schwartz, on Abydenus, Alexandros 88, and Berossus, in the Pauly-Wis- 
sowa Real-encyclopddie. 



ASSYRIAN HISTORIOGRAPHY 65 

ties for securing the truth, all the greater must be our suspicion 
that they have concealed the truth when it was not to the advan- 
tage of the monarch glorified. Only when we have applied these 
principles in detail to the various documents can we be sure of 
our Assyrian history and only then shall we understand the men- 
tal processes of the Assyrian historians. 



ABBREVIATIONS 



Abel-Winckler 

AJSL 

Amiaud-Scheil 

AOP 
BM 



Budge 
Budge-King 

G. Smith 
Harper 

JA 

JRAS 

KB 

KTA 

L 

Le Gac 

MDOa 

Men ant 

NR 

OLZ 

PSBA 

R 

Rasmussen 
Rogers 

Rost 

RP 

RT 

S. A. Smith 

Smith-Sayce 

TSBA 

Ungnad 

ZA 



L. Abel, H. Winckler, Keilschrifttexte, 1890. 

American Journal of Semitic Languages. 

A. Amiaud, V. Scheil, Les inscriptions de Salmanas- 
sar II, 1890. 

H. Winckler, Altorientalische Forschungen, 1893£f. 

British Museum number; special collections are 
marked K., S., Rm., DT., or by the year, month, 
and day, as 81-2-3, 79. 

E. A. W. Budge, History of Esarhaddon, 1880. 

E. A. W. Budge, L. W. King, Annals of Kings of As- 
syria, I. 1902. 

G. Smith, History of Assurbanipal, 1871. 

R. F. Harper, Assyrian and Babylonian Literature, 
1901. 

Journal Asiatique. 

Journal of the Royal Asiatic Society. 

E. Schrader, Keilinschriftliche Bibliothek, 1889ff. 

L. Messerschmidt, Keilschrifttexte aus Assur, I. 1911. 

A. H. Layard, Inscriptions in the Cuneiform Char- 
acter, 1851. 

Y. le Gac, Les inscriptions d'Assur-nasir-apal III, 
1907. 

Mittheilungen der Deutschen Orient Gesellschaft. 

Menant, Annales des rois d'Assyrie, 1874. 

A. H. Layard, Nineveh and its Remains, 1851. 

Orientalistische Literaturzeitung. 

Proceedings of the Society of Biblical Archaeology. 

H. C. Rawlinson, Cuneiform Inscriptions of Western 
Asia, 1861ff. 

N. Rasmussen, Salmanasser den IPs Indskriften. 

R. W. Rogers, Cuneiform Parallels to the Old Testa- 
ment, 1912. 

P. Rost, Keilschrifttexte Tiglat-Pilesers, 1893. 

Records of the Past, Ser. I. 1875ff.; Ser. II. 1889ff. 

Recueil de Travaux. 

S. A. Smith, Keilschrifttexte Asurbanipals, 1887ff. 

G. Smith, A. H. Sayce, History of Sennacherib, 1878. 

Transactions of the Society of Biblical Archaeology. 

A. Ungnad, in H. Gressmann, Altorientalische Texte, 
1909. 

Zeitschrift fiir Aseyriologie. 



(66) 



UNIVERSITY OF MISSOURI STUDIES 



SOCIAL SCIENCE SERIES 



VOLUME I 

? 

The Clothing Industry in New York, by Jesse E. Pope, Ph. 
D., Professor of Economics and Finance, pp. xviii, 3.40. 
1905. $1.25. 

VOLUME II 

1. The Social • Function of Religious Belief, by WrLLiA^r 
Wilson Elwano, Ph. D. pp. viii, 100. 1908. $1.00. 

L'. The Origin and Early Development of the English Uni- 
versities to the Close of the Thirteenth Century, by 
Eakxest Vancourt Vaughn. A. M., Instructor in His- 
tory, pp. viii, 147. 190S. ?1.00. 

3. The Origin of the Werewolf Superstition, by Caeoltxe 
Taylor SxrrwARX. A. M., Ph. D., Assistant Professor of 
Germanic Languages, pp. iv, 32. 1909. 

4. The Transitional Period, 1788-1789, in the Government 
of the United States, by Frank Fletcher Stephens, Ph. 
M., Ph. D., Instructor in American History, pp. viii, 
126. 1909. $1.00. 

VOLUME III 

1. Assyrian Historiography: A Source Study, by Alhert 
Ten Eyck Ol-mstead. Ph. D., Associate Professor of 
Ancient History, pp. viii, 65. 1916. $1.00. * 



T -BR VRY 



"CHBik 




LrJ 



RETl 
TQi 

LOAr 
H 



6-moi 



AEC 



HnMi: liCE 

RETURN TO the circulation desk of any 
University of California Library 

or to the 

NORTHERN REGIONAL LIBRARY FACILITY 

BIdg. 400, Richmond Field Station 

University of California 

Richmond, CA 94804-4698 

ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 

• 2-month loans may be renewed by calling 
(510)642-6753 

• 1-year loans may be recharged by bringing 
books to NRLF 

• Renewals and recharges may be made 
4 days prior to due date 



DUE AS STAMPED BELOW 



APR 1 7 2003 



&^ 



FOR/ 



DD20 15M 4-02 



<ELEY 




'^f^h 



UNIVERSITY OF CALIFORNIA LIBRARY 



^ GENERAL LIBRARY - U.C. BERKELEY 



BQDQEfiSS31 




,.,. ..,,f,i.-'i»/ri 







. ,, *^ , .>v»-.ji ->' '5'- It: